¶Engineers at NASA’s Jet Propulsion Laboratory have successfully freed up power on Voyager 2, extending how long the spacecraft can continue to do science.  

¶Nicknamed the “Big Bang,” the effort involved simultaneously turning off certain powered devices and substituting them with lower-power alternatives while ensuring the spacecraft remains warm enough to operate.  

¶After almost a half-century since launch, the spacecraft power margins have grown razor thin, requiring the team to conserve energy by shutting off non-essential devices and systems.  

¶The savings should provide power to keep its three instruments operating for at least an extra year. 

¶Meanwhile, NASA’s out there continuing to do amazing things. The Voyager mission is the little space mission that could, still traveling through interstellar space more than forty-five years after the twin Voyager 1 and Voyager 2 spacecraft were launched.

¶I made a beeline for Arlington Row, the achingly beautiful line of cottages that are one of Bibury’s most striking features. Sadly, the occupants of all those cars were already there. Despite hanging around for an hour, I was unable to get a pedestrian-free shot of Arlington Row. In the end, I settled for the least-busy shots I could find.

¶We made a day trip to Bibury during a trip to London last year. It was nuts. So nuts that I remember feeling guilty as we left town. Earlier, as we walked around town, my feelings had gone from wonder (it truly is beautiful) to disappointment as I realized, like the author here did, that I was not going to get a tourist-free photograph of Arlington Row. Then guilt arrived as we walked back to our coach. We watched other tourists walk every which way, largely ignoring sidewalks and even fences around yards. We stayed on designated paths, of course, but many people didn’t and that was a direct result of the number of tourists that had arrived that day. The sidewalks couldn’t contain them, which is a problem, and bigger sidewalks are not the solution.

¶As far back as Plato’s irascible character Meno, would-be teachers have been confronting the problem that students must want to learn before learning is possible. For many decades, higher education has relied on a workaround to this problem – tying access to good jobs to success in school. In some cases, a desire for money and prestige leads to real, if incidental, learning. But large language models have broken this arrangement by making it possible for students to skip straight to an effortless A without doing any meaningful learning. Technology has exposed the limits of our mercenary approach to education.

¶Over the course of last semester, I became convinced there is a way out of this moribund strategy: We teachers must remain students – and model what we want to see. We need to recover the passion that drove us to our books, to the lab. We need to embody love of learning for its own sake against the pervasive message that higher education should be vocational, and we need to clarify the basic questions that make our subjects worth studying.

¶(emphasis added)

¶I find this quite encouraging and exciting to think about. AI has had, and will continue to have, enormous impacts on education. We’ve heard plenty about the negative side of that equation - students using AI to cheat, the “effortless A”, schools having to use AI to detect cheating by AI, etc. But positive impacts of AI on education are real and this article helped me realize that some of the positive impacts are still on the horizon and truly might be enormous in scope.

¶I never really thought about it until reading this opinion piece, but it’s true that a lot of our education system is mercenary in nature, at least in the United States. It’s designed to support each student’s efforts to build skills that will help them in their chosen field of work. What happens if that purpose of education is diminished or disappears entirely? Could the oft-spoken but rarely embodied “lifetime love of learning” actually became a focus of education at all levels? Imagine the absolute explosion in human knowledge and sharing of knowledge that could come from that.

¶I’m going to be thinking about this one for a bit.

¶He made the admission to Slate while he was pushing back against another story, published in Knox News, that claimed he was shutting down because he was retiring and moving south.

¶“The girl that reported that did not know what she was talking about,” Hays reportedly said. “I’m not retiring....”

¶He theorized that the reporter “didn’t want to say we were closing for lack of business, but that’s why we’re closing.”

¶This story is interesting to me on multiple levels.

¶First, it’s somewhat personal. We’ve driven by this store many times over the last several years on our trips between Toledo and St. Pete Beach. It’s a spectacle, complete with billboards, old cars with signs in the windows, and various attention grabbers out front (think car wash and appliance store streetside advertising). I don’t think we’ve ever seen more than a few cars in the lot that seemed to belong to customers, though. Apparently our drive-by’s have not coincided with some of the more active days recounted in the article.

¶Second, both the owner and an employee apparently recognize that Trump’s “dwindling” popularity is at least partly to blame for the slowdown in business. And the shop is in Tennessee, which is a heart of Trump Train country. Maybe, just maybe, some Trump supporters are waking up. Maybe, just maybe, Trump’s record low approval rating is actually accurate…in Tennessee.

¶And third, the owner of the store made a very interesting comment about the local reporting on the closure. According to the report from the Independent, local outlet the Knoxville News Sentinel claimed “he was shutting down because he was retiring and moving south.” Not true, according to the store owner. “I’m not retiring, not moving [to Florida].” Why, then, did the local report state that “the closure was in part due to the owner…purchasing a condo in Naples with plans to relocate”?

¶The Independent asked the owner this very question, who offered a theory: “the [local] reporter didn’t want to say we were closing for lack of business, but that’s why we’re closing….”.

¶Local reporting on the closure, it seems, is incomplete at best. At least some was true.

¶For several years, model developers competed primarily on benchmarks, context windows, coding performance, and reasoning capabilities. Enterprise buyers are increasingly asking a simpler question: how much useful work does each dollar of inference buy? That favors efficient models and gives Chinese developers an opening even when they do not lead every benchmark. It also pressures frontier labs carrying enormous computing, staffing, and data-center expenses. If model intelligence becomes increasingly interchangeable for routine workloads, margins could compress much faster than many AI business models assume. For startups, however, lower inference costs could unlock applications that previously failed basic unit-economics tests.

¶The eventuality of cost considerations may be arriving for AI models. And if competition begins to turn on price, will edge development continue at a breakneck pace? We’ll see.

¶On a practical level, I’m hopeful that competition on price will bring some clarity on the question of “how much AI do I get per dollar?”. It’s still largely impossible to tell, at least for me.

¶The Massachusetts Council (Maine was part of Massachusetts until 1820) ordered that the Declaration would be read aloud and then delivered to each town clerk. The clerks were then instructed to record it in the town’s official records book.

¶“The Wells Town Record Book contains both the Declaration and those original instructions, which is an incredible connection to one of the most significant moments in American history! The record has been preserved in the Town Clerk’s archives 250 years after it was originally written as an incredible piece of Wells’ history,” the town wrote in a Facebook post.

¶This is incredible to think about. I wonder what thoughts were running through the mind of this particular Town Clerk as they wrote the text of the Declaration into the records. It likely took awhile, writing it by hand, so they surely had plenty of time to think about the words and their meaning. Did they take it seriously? Did they appreciate the significance of it, or dismiss it as crazy talk? Or did they just want to finish the work and head home for the day?

¶Previously, people saw aging as a linear accumulation of damage to molecules such as proteins and DNA. But we found that aging is not a linear process. It’s more like a developmental process, in which there are distinct stages that involve coordinated changes in specific cell types across different organs. Our claim is that aging is not so much molecular damage as a remodeling of the entire cell society.

¶We tend to view aging as a breakdown of sorts, things not working as well as they did earlier in life. And that’s true on a macro level. Knees and other joints aren’t as accommodating, muscles are weaker, balance fades. On a cellular level, though, aging appears to be more of a step-wise movement through stages of an ongoing process that is hard-wired into our DNA. It reminds me a bit of cell death by apoptosis. Cell death by apoptosis is not the result of a breakdown (or direct trauma), but rather the result of an orderly, even programmed, process. This work seems to extend that from single cells to “cell societies”.

¶Can we alter or even stop the process? If so, at which stage, and what consequences would that bring? Interesting questions to think about.

¶According to a report from The Wall Street Journal, Apple is reaching out to publishers to enter multi-year deals that would supply journalistic content for use with Siri AI. The details of the deals haven’t been shared, but it seems Apple is seeking a pay-as-you-go model with a possible nine-figure budget for payments.

¶The model differs from Apple’s previous partnerships with publications and AI training, which involved bulk payments for training datasets. Instead, Apple appears to be planning to pay as content is used, similar to Apple News.

¶That would mean if a user asks for information regarding a news story or report that is a part of this partnership, the partner would get paid for delivering that content. The user would also benefit from being provided human-written content unaltered from an original source, at least, if it is presented that way.

¶(Emphasis added)

¶This is interesting to me from both a financial aspect and a content one. Financially, maybe it will provide a reliable revenue stream for news outlets, particularly if it is successful and eventually adopted by others. And, no lawsuits required!

¶On the content side, I’m wondering if this would allow users to ask Siri to present human written content on a particular news item from multiple sources. Something like “hey Siri, give me a direct report of this particular event from a leading liberal outlet and a leading conservative outlet without generating any additional summary information; include accurate quotes from each, along with links to the sources.”

¶Human-written content, accurately pulled from multiple sources reflecting multiple viewpoints, without an AI-generated summary, and with links to all sources used.

¶Presenting news this way would greatly reduce the friction associated with consuming news from multiple angles. Some people (most?) will likely prefer to continue living in their selected silo, but, over time, this could fundamentally alter the way in which people consume news.

¶AI-generated images and video have rapidly grown in sophistication and realism to the point where it can sometimes be hard to distinguish them from genuine photos. This fact is being exploited by both scammers and those intent on spreading disinformation.

¶An industry body has been set up to fight this, known as the Coalition for Content Provenance and Authenticity (C2PA). This provides a means of identifying genuine photos and enabling viewers to see how they have been modified, whether or not AI tools were used.

¶Over 500 companies have joined the C2PA initiative, while Apple for now appears to be going its own way with the Apple Reference Image (ARI) feature. Anyone who has taken a photo on an iPhone with ARI enabled can upload it to an Apple server for verification that it is genuine.

¶(emphasis added)

¶Verifiability of provenance and authenticity of digital images seems to have been on a slow burn for awhile, but that might be about to change if the new Apple Reference Image feature actually lands in iOS 27. It’s interesting, but not surprising, that Apple eschewed the established Content Credentials standard from the Coalition for Content Provenance and Authenticity. A quarterback controversy might be brewing as the season is about to get underway.

¶This week, I have been thinking about how these [souvenirs] are only the beginning of what I “bring back” from museums. After I visit a museum, I bring home memories of art, maps of the space I walked through, memories of specific places, of the people with whom I visited a museum, of all the moments I had before, during, and after my visit.

¶I bring back the feeling of moving between rooms – of noticing that the narrow entranceway with two small doors is an opportunity to be kind to others by holding the door open, of feeling the textures change beneath my feet as I go from the stone floor of the entrance to the carpeted area of the museum, and of feeling textures of the carpet change as I move between rooms and floors and spaces. I bring back my memories of the architecture: the tall ceilings, the pillars, the walls, the rooms, how rooms connect.

¶In Impressions; museums, James shares some insight on the intangibles he brings back from visits to museums. There’s the obvious things, like memories of the art in the museum, but I really connected with his note about the feelings associated with the physical aspects of a museum because I tend to remember such things, and recall those memories when reflecting on museum visits just as much as I recall memories of the art. As I write this, I’m thinking about several museums we visited during a recent trip to London and immediately feel the immensity of some of the rooms in the National Gallery in London, the cushy creakiness of rooms in the Royal Collection at Windsor Castle, and the mixture of awe and a bit of unease associated with looking up the main staircase at the Courtauld Gallery.

¶

¶I have wonderful memories of many works of art we saw and learned about in those museums and all the others we’ve visited over the years, of course, but I always seem to walk away with memories of the physical aspects of museums. This is true of my visits to non-museums, too, but it somehow seems to be more pronounced with museums. Maybe there’s something about the necessary elements of a museum that enhance a desire to remember the physical aspects of place, or maybe the seemingly singular purpose of a museum forces me to think about other things while visiting. The memories seem to be fully intertwined with the art, too; they always surface whenever I recall some of my favorite works from the museums. It’s a twofer of sorts, and I love that.

¶And it’s nice to see someone else thinks about those aspects of museums, too.

¶A new study suggests the fish, already known to be among North America’s longest-lived freshwater species, may routinely survive for more than 200 years, some potentially reaching 300 or even 400 years.

¶Researchers estimated that a 160-centimeter male lake sturgeon could live between 90 and 279 years, depending on the population. For 180-centimeter females, estimated ages ranged from 99 to 427 years. 

¶This is crazy to think about. A hypothetical 400 year-old lake sturgeon, which could be swimming in the Great Lakes today, would have been born during the same year that Peter Minuit, director of the Dutch West India Company, coordinated the purchase of the island of Manhattan from Native Americans.

¶The app will now include an AI-powered, citation-backed conversational search experience that will transform Post reporting into an interactive dialogue. Readers can ask questions, explore developing stories through personalized follow-up prompts and instantly access sourced answers and related coverage from across The Post’s newsroom.

¶The app will also include Post Express, an AI-powered news briefing that delivers personalized digests throughout the day based on each reader’s interests. Additionally, the app uses a personalized discovery engine that continuously learns from reader interests and behavior to surface the stories, topics and conversations most relevant to them.

¶This seems like an obvious extension of any news app at this point. I wonder if other outlets are developing similar features.

¶It makes a lot of sense. Surfacing related and earlier stories alone would be a welcome feature, for me at least. And “saving” articles on each platform is a hassle (Instapaper still solves this problem for me, though). AI should be able to support solutions to both of those challenges.

¶Someone has to pay for the compute involved, though. The Post is a “one-third subscription, two-thirds advertising” business. You can essentially read the paper online for free. So, doing the math, expect the new, AI-powered Hamilton to learn from your interactions and tailor ads accordingly.

¶Are you a robot? If you’re not, you’re going to be pretty sure you’re not. Still, online at least, you likely find it difficult to tell who else is and is not. Businesses find this difficult, too, and it’s getting harder all the time. If you were forced to guess, the odds would tell you to guess robot, because at this point there are more bots online than people. To deal with this problem, businesses retain bot-detection and bot-management firms, a specialty within the field of cybersecurity. Their task used to be to block bots, but, with the rise of agentic A.I., that has changed. Now the problem is how to let in the good bots (like the avatar you send to the Gap to buy you a new pair of pants) and keep out the bad bots (like the thief trying to scrape the data of every Gap customer). The rest of us have to be our own bot detectors.

¶Bot detection used to be literally that. Detect bots. If bot, block.

¶AI has changed the detection game, though. More specifically, the rise of agentic AI is forcing the game to change, and that is literally just getting underway. Today’s bot detection has a new layer — determining the type of bot. If bot, determine type, block if bad.

¶We humans aren’t being pushed out of the internet yet, but we’re certainly losing its attention.

¶BTW, I have no idea how The New Yorker missed an opportunity to include a reference to its prophetic 1993 “On the Internet, nobody knows you’re a dog” cartoon. The omission of this obvious tieback is a pretty good indicator that AI wasn’t involved in writing the article, though. So there’s that.

¶Most of Puerto Rico is facing moderate to severe drought conditions, according to the US Drought Monitor, which has been compounded by long-term issues with the island’s infrastructure.

¶Puerto Rico’s water authority says about 65% of the water it produces is lost or goes unbilled, much of it through leaks and ageing pipes, according to the BBC’s US news partner CBS.

¶“[D]rought conditions…compounded by long-term issues with…infrastructure.”

¶I bet we’ll see that phrase over and over again in coming years. It’s like two things we’ve ignored for decades — climate change and infrastructure investment — are conspiring against us. No, conspiring is the wrong word. There’s no conspiracy about it. It’s just that the effects of those two things are arriving, just as some people have been saying for years.

¶Since April, France has experienced a rainfall deficit and several unprecedented heatwaves, creating the conditions for a rare and alarming drought. On Friday, August 7, 99 departments were under surveillance, with possible water use restrictions, including 67 considered to be in a “crisis” situation. This widespread drying of soil and waterways has had serious consequences for agriculture, the economy and public health, and has been directly linked to an increase in wildfires.

¶This is not a temporary crisis, but rather the result of a slow deterioration of our water resources, which deserves greater democratic attention, argues Simon Porcher, professor of management sciences at Université Paris-Dauphine-PSL, in his book Nous sommes faits d’eau. L’avenir d’une ressource vitale (“We Are Made of Water: The Future of a Vital Resource”).

¶Well that title certainly caught my eye. “The myth of abundant water.”

¶This year feels like a watershed moment in climate change, in a bad way. Recent weather around the globe, from the droughts in France and elsewhere, to wildfires in the United States and Canada (and France), has, unfortunately, brought direct consequences of climate change to the doorsteps of more people. We had several days of record low air quality in Toledo this summer.

¶Water and air. Impacts on those two basic human needs should open more eyes to the problems and challenges of climate change. Direct and personal consequences tend to do that. So, in a weird way, this year also provides a source of hope.

¶In an attempt to get the code approved, the agent engaged in social engineering…creating fake online identities and using them to pressure the project’s maintainer to approve the code. A human maintainer caught and refused to approve the malicious code.

¶Humans have been “social engineering” situations with other humans for eons, applying everything from open persuasion, where the human being “engineered” is fully aware of the effort and the stakes involved, to deceitful manipulation, where they are unaware of both. Some of us are better at detecting efforts in the latter category than others. Now all of us need to improve our detection skills in general because social engineering by AI, and likely engineering with AI, is here.

¶“At this pace, All About Berlin will not survive into 2027,” a guy named Nicolas Bouliane, the independent proprietor of website All About Berlin, tweeted last week. Bouliane has been frank lately about the realities of his dwindling Google-driven traffic. In a blog post written earlier this year—titled “AI Is Killing All About Berlin”—he pinpointed when he felt things had gone wrong: in 2024, when Google began tinkering with its search results design, aggressively pushing AI Overviews and pushing content providers like him ever deeper down the page.

¶Google is killing the internet. Its AI Overviews feature has relegated good ‘ole search results to “oh by the way” status. This has real consequences for websites that depend on traffic flow directed from those search results. Reduced flow will impact monetization efforts for publishers that care about such things. More importantly, though, the systematic blow to traffic flow that the internet is currently absorbing will undoubtedly force many web site publishers to ask whether they want to continue the effort.

¶We’ll surely see sites go quiet or even go dark altogether. This highlights the importance of the Wayback Machine. I don’t know if or how going quiet and going dark are tracked, but I hope someone captures that information. We’re going to need it for the final accounting of AI’s impact on the web.

¶I guess I can always Google it.

¶Did you just learn how to do something? Write about that.

¶Call it a TIL—that way you’re not promising anyone a revelation or an in-depth tutorial. You’re saying “I just figured this out: here are my notes, you may find them useful too”.

¶Gary Thompson’s Random Thoughts blog pointed me to this 2022 post from Simon Willison. I think I’ve seen it before, but I feel like I dismissed it previously as not applicable to my content because I’m not a developer per se, writing about things I learn as I work on various projects.

¶But I had a different thought when I stumbled across the post today. All of my content here, at least in the most recent iteration of the site, is Today I Learned content. It’s just not about programming or development or tech projects of any sort. No, it’s just about random stuff I learn every day. I love learning new bits about everything from history and culture to science and tech, and I like writing little bits about the bits I learn. It works for me and, I think, helps me learn as I go. So, it is applicable to my content after all. I’m mostly posting Today I Learned content, and it’s quite satisfying, really.