I think part of the reason Ford is having issues is that their EVs are or are going to be gigantic.
Ford’s been riding the trend of bigger and bigger trucks and SUVs such that they’ve built themselves out of sedans and smaller trucks that are probably better suited for EVs.
Indeed, I think the trend of the past few years of larger SUVs and trucks actually has given folks the wrong expectation of what cars should be as we enter the EV-era.
The car company says it isn’t backing off its next-gen EVs, but it is pausing some big factory projects, including a plant in Kentucky.
Just wanted to say: this is a really good article.
The article is a great snapshot of where we are in SEO, with insights from the start many decades a go. Great quotes from interesting characters, too.
As the public begins to believe Google isn’t as useful anymore, what happens to the cottage industry of search engine optimization experts who struck content oil and smeared it all over the web? Well, they find a new way to get rich and keep the party going.
It’s a gadget designed for interacting with large language models, not apps, and for talking instead of typing. But it’s not yet entirely clear what you’re supposed to use it for.
I’m not so worried that this is a new kind of device, trying new forms of interaction and design. And kudos for being gutsy to use LLM so early in the evolution of the tools.
But already folks are pointing out LLM errors in the demo. And I’d like to point out in the launch video, Bethany, Humane’s woman co-founder, never speaks with the device [my experience with Siri and Amazon Echo showed that women have a heard time being understood by voice input tools, let alone those in the over-55 bracket]. And do we really want some device blaring out results in public, results that might be quite personal, instead of being respectfully silent and replying thru a headset?
Again, I welcome new devices that try new things and attempt to change our expectations of what our tools should do and how we should interact with them.
And I totally agree that we’ll really only see how useful this is after it’s out in the wild and being used.
But, will it go the way of smartwatches (well-received for many reasons) or the way of smartglasses (well-received in particular use cases, but by no means a consumer device)?
Humans are really good at coming up with better ways to create things. And humans are also very good at pointing to anything new as the end of the world as they know it. And those folks are not far off their prediction, but not how they think.
Writing will make you styoopid Did you know that Socrates was against writing? Writing wasn’t a newish thing in his day. Nonetheless, he didn’t think it was an effective way of communicating. And he worried that folks would become forgetful.
But writing has been central to humanity’s growth and dissemination of knowledge and in the sharing and enjoyment of stories. Can you imagine a world without writing? You would have to meet up with me for me to tell you what I’m writing here.
Derivative drivel I’ve been following how folks have been reacting to generative AI for making images, text, music, and audio. The worried ones are concerned about copyright, plagiarism, thinking effort (familiar, eh, Socrates?), and such.
Yet, if you think of the process of writing, there’s extensive reading (indeed, the best writers are voracious readers), there’s the exercise of learning styles from existing sources, there are editors and proofreaders to help the best writers be better. And we know that plagiarism and ghost writing happens.
So what’s the fear with ChatGPT? The system is trained on a broad range of material – just like anyone who reads everything. The system synthesizes all of that into an output – just like anyone who writes.
Indeed, the article quoted below from the NYT is one of the few very well-balanced explorations of the impact of generative AI, in this case, on college admission essays.
A.I. chatbots could facilitate plagiarism on college applications or democratize student access to writing help. Or maybe both.
Good writing is always good writing One thing that no one seems to point out in all the articles I have read: no matter WHO writes the text, krap writing is still krap writing.
In a previous post where I used ChatGPT for programming, I mentioned that ChatGPT is a krappy programmer, but was helpful to accelerate my programming tasks. In the end, I had to know how to program for ChatGPT to help me.
The same is for any writing: you need to be a good writer to turn the ChatGPT text into good text.
_____ will democratize ____ When blogs came out, folks said that anyone could become a journalist, everyone would be writing online and there would be an explosion of text online.
But in the end, the best blogs were written by writers and journalists.
When podcasts were reborn, folks said everyone would be able to create audio channels of all sorts. But the podcast that brought podcasts from the dead, Serial, was produced by a long-time radio crew. And all the best podcasts are made by professional audio teams.
Good writing and radio is made by folks who know how to create, do the effort, and produce good writing and radio. Generative AI will not displace them.
As one of the students mentioned in the NYT article says:
But she found the responses too broad and impersonal, even after she gave it details about her extracurricular activities like teaching dance classes to younger students.
“I feel a little more pressure to create, like, this super unique, interesting topic,” Ms. Vakharia said, “because a basic one these days could just be generated by ChatGPT.”
In short, she knows she still needs to put in the work to write a good essay.
Back in May, I mentioned a project that I was working on, inspired from a line in a book, leading me to hack solar deck lights.
I just want to say that I finally put it all together. I made two candlelight-like motions-sensing solar-charged pieces, and one pice that also had LEDs for blinking like firelflies.
An aside: Fireflies? A few years back, I made a gizmo that would turn on at dusk and blink a firefly pattern. Alas, it was only a single LED, and a very crumple circuit.
For this latest project, I also wanted to do LEDs flashing, but chose to do six lights, randomly flashing, all to the same firefly pattern.
I will admit that I used ChatGPT to help me figure out the blink patterns and then program it. Alas, I first asked ChatGPT to program in CircuitPython, so I could use my Raspberry Pi PICO. But something wasn’t right and I think I fried two of them. SO I chose to them just do it with an Arduino Nano clone. ChatGPT was kind enough to convert the code (tho, don’t tell it that it did a mediocre job and I had to clean up the code).
In the end, the third part of the project was a candle-light motion-sensing solar-charged piece that also blinked for fireflies.
The other light-driven part Now that I had the three solar-power parts, I filled up a flowerpot and spread wildflowers around (see image, right).
And waited. Until the flowers grew.
Now I have a pretty flowerpot with flowers that draw life from light and three candle-light electronics that draw life from light (see upper left image for dusk and day views).
A while back I read an interesting article that placed hobbyists making things (Level 1 – 1-100 units) on the same axis as big manufactures (Level 3 – 5000+ units) [as an aside, the writer now has a robust 7+ person company making tools for Level 2 manufacturing].
I’m definitely in the Level 1 side of things. I often make just one of something.
But in the world of electronics, often you end up with multiple boards for really cheap or you want to make a few dozen of one board for your own use. Assembling things and soldering components is a pain at that scale, especially if you want to use the more compact surface mounted components, and especially if those surface mount components are really teeny.
Easy Bake I find being able to use free tools to design a circuit board, easily print the boards, and then use cheap components to build your own circuitboard as a kind of superpower. The range of creations out there that anyone can design and build is wonderful. (can I interest you in #badgelife?)
What a lot of folks do when they have a bunch of boards is either use a hot gun to solder a boardful of components, or do as the pros do: use an oven.
There are hacks and such of folks manually reflowing the solder, following a solder reflow profile of different temperatures. But the ease of tools-beget-tools means there are controllers one can use to automate the oven process.
Unexpected purchase In the process of moving house, my mother had a toaster oven that no one needed, so I claimed it specifically to make a reflow oven. [🤔Hm, I wonder if I would have ever started along this path to making an oven if I didn’t have this free toaster].
I then picked up a controller from a great guy, who, by the way, went thru the three layers of manufacturing in the past seven or so years – he’s now a Level 3. Indeed, the origin of his reflow oven controller was from his time as a Level 1, before he bought a proper oven as he stepped up to Level 2.
I ripped out the guts of the toaster, rewired it to be directed by the controller, and ran a test run.
Alas, seemed like my oven (in green) couldn’t keep to the desired temp profile (in pink).
I did some reading. Saw a lot about oven volume and element strength, insulation.
So I figured I need to insulate the oven big time and get things in order.
Interesting but a non-experiment I wrapped the insides in a insulating blanket, plugged up some holes, added some reflectors. And then, excited, ran the test run again.
WTF. No better than uninsulated. Then I see that the controller maker never insulated his oven. And thinking of it, the time frame is so short, probably doesn’t make a difference (which of course the data supports).
Then I wondered if my two elements just couldn’t hit the temps, even though they should.
A test run at really hight temps showed that the elements and oven are more than able to reach the temps in time, albeit, not matching the desired curve.
Then, in what are those, ‘fuck, why did it only occur to me now after all the work I put in?’ moments, I realized I could tweak the controller to hit the needed reflow temps by just raising the desired temps in the controller profile.
Worked. Now, even though (left below) the pink measured temp don’t match the green desired temps, I am hitting the right temps for my reflow.
Indeed, I was able to match the required temp curve for my solder paste reflow profile (right below).
Keep calm and make something Of course, all this was leading up to a board I wanted to make.
As this was the first board I made in a reflow oven, I learned a lot in the process: This was my first proper project to use solder paste, meaning getting a stencil and making a jig to hold things. The components were the smallest I’ve ever used. The board was double-sided, so needed a tricky two rounds of reflow. The chips had closely spaced pins, so needed to clean that up – rework. And I had some thru-hole parts that had to be added, that were easy peasy.
And, yes, the product worked as designed. And that’s why I made it myself as there was no one selling something similar. Fortunately for me, someone had designed this one and graciously made the design available.
What’s next? Funny thing is, when I bought the controller a while back, I had a lot of ideas on things I wanted to make. But now, my head is elsewhere and the oven has not been used since the run of a few copies of the board above.
Hm, could also be that the price of just having someone else assemble the boards is falling fast, and, well, after buying all the oven and components and boards and all, and putting in the time needed, this board might have been cheaper just having it assembled for me. This also means perhaps I now have a higher hurdle to make it worthwhile to use the oven.
Let’s see.
From China with Love As with most purchases of boards and components from China, when you want only one, you end up getting more than 10 of everything (soo inexpensive!). Therefore, I still have a handful of extra unassembled boards. I might just complete all those and sell them (at cost, most likely, as they were not designed by me).
And for the boards I had in mind way back when, I think I will have to balance the cost of me assembling them verses someone else in China doing it.
Looking back and ahead Two things are constant in the maker world.
One is, why buy it when you can make it for more?
Two is, often it’s the journey that counts, not the destination. In the process of doing, you learn, you share, you discover new things that open new doors.
I did spend a pretty penny on building that reflow oven and making one copy of that board. And who knows if I will ever make that money back in savings making my own boards, in these days of cheap assembly. Nonetheless, I do feel the whole journey was fascinating, granted me a whole level of abilities, and opened up a range of new things I can do.
Let’s see where that path of this journey takes me next.
For two decades, Google Search was the invisible force that determined the ebb and flow of online content. Now, for the first time, its cultural relevance is in question.
Get in, loser, we’re going back to Web 1.0. We have the opportunity to get out from under the algorithms. So maybe it’s time to think about what a web of people looks like now.
The wicked witch is dead, Jim In the past year, Elizabeth has articulated so well the unravelling and dysfunction on the web. She compares that to the early days of the web (say, pre-2005) where the ubiquity of phones and feeds didn’t exist. And she rues the dominance of the algorithms from behemoth controllers of the internet.
I was there on those early days (this blog has a lot of commentary from back then). We were excited about closer social connections across space and time, of democratizing content creation for all, of breaking the shackles of powerful incumbents.
Alas, we changed one set of shackles for another. And, as anyone knows who builds selection algorithms, be careful what you select for.
Phone it in And back then I was an ambassador to the new world of carrying a smartphone, hovering up your life thru notes, pictures, and videos and sharing them with the world (indeed, part of a team building such an app). I also took it to the next level, promoting (actually, also trying to build) the brave new world of finding, mixing, and publishing social feeds online.
Ah, the hope of it.
Dyst-hope-ia Well, you know where that got us.
Elizabeth echoes so much of where I’ve been going. I got off of all social media. Hang out now in bulletin-boards, aka Discord. I mostly use my phone for group chats and looking up information. And lots of phone calls.
We’re not alone Elizabeth is not just articulating a personal feeling she senses, but has tapped into the zeitgeist. Just check out the enthusiasm in the comments of her article (OMG, interesting and considerate comments – how early-days Web 2.0).
There’s hope for where we go from here.
I, for one, have been trying to get back my blogging mojo from the early days of this blog. I think I also need to get better at reading other folks’ blogs.
Maybe the future of the web is to go back to the past of the web. Indeed, Elizabeth exhorts folks to avail of the new tools that make it easier than ever to build their own internet, with their own friends, and connect to the people that matter.
Casey Newton from The Verge has an interesting ‘state of note-taking apps’ article. Prompted by the arrival of generative AI, he looks back at his use of note-taking apps and ponder the use of actually thinking with them.
the arrival of generative artificial intelligence could make the tools we use more powerful than ever — or they could turn out to be just another mirage.
I’ve been a daily user of Evernote for something like 15 years. And even before that, I had some sort of database or collection of digital notes of some sort.
I wouldn’t say I use Evernote for great insights, since, as Casey points out, note-taking apps are not really geared for that. I do use it as an outboard memory, saving snippets of things I later search for.
I also use it as my lab notebook, for all my maker projects (and perhaps some other projects I want to capture process for). Tho, a few months (and projects) had passed before I realized I should be documenting my work like I did on paper as a grad student and post-doc.
AI? Evernote has added some sort of AI summarization feature, which I have still to try out. I suppose I am not so much interested in a summary of a single note, but perhaps a summary of multiple notes.
Tho, as Casey points out, while wishing for something similar, the nature of AI summaries these days leaves much to be desired.
In the end, Casey realizes that expecting a piece of software to do the thinking for you is the wrong path to try. “Thinking takes place in your brain,” he says. And “it is a process that stubbornly resists automation.”
Indeed.
Tool use He then refers to Andy Matuschak, a thinking on notes and note-taking:
“The goal is not to take notes — the goal is to think effectively,” Matuschak writes. “Better questions are ‘what practices can help me reliably develop insights over time?’ [and] ‘how can I shepherd my attention effectively?’”
In a way, that’s how I approach any tool – how does a particular tool fit into the habits I have to achieve something? In the case of notes, I am constantly evaluating the value of Evernote and other options for how I collect, save, navigate, and recombine my morsels of information (heck, Evernote is quite good at screwing things up and forcing all their users to question Evernote’s value, haha).
Right now, after so long using Evernote, I have a way that works for me. And there are only a few tweaks I’d do to Evernote to better fit how I use it (alas, many are features that were lost – say, support for GIFs, or modern document and card scanning).
Future? What would I want AI to do in Evernote?
Not sure. Of course, some sort of overlay to organize info and find connections, rather than a co-writer, would be useful (maybe like the nascent NotebookLM Google?).
I really don’t want it to try to be smarter or as smart as me.
BASAAP is always much better for an assistive digital tool.
This article got me thinking of Galt’s Gulch in Rand’s Atlas Shrugged, or Freeside in Gibson’s Neuromancer: places the wealthy run off to to escape mainstream society. Indeed, what this article talks about isn’t something that echoes these fictions, this is an actual escapist effort in our own reality.
Reid Hoffman, Marc Andreessen, and Laurene Powell Jobs reportedly all contributed to the venture to build a new city with ‘tens of thousands of new homes’ and ‘orchards with over a million new trees.’
“Impact of privilege and isolation” I was trying to remember other similar books that had this or something similar as a plot tool, such as in Sagan’s Contact or Stephenson’s Snow Crash. So I had a conversation with ChatGPT about this and then had it summarize a few for me:
Certainly, the concept of wealthy individuals or groups leaving mainstream society to establish their own secluded enclaves is a recurring theme in literature. Here are a few more examples of such settings from various novels:
“The Diamond Age: Or, A Young Lady’s Illustrated Primer” by Neal Stephenson: In this novel, there are gated communities known as “phyles,” which are separate cultural and societal groups. Each phyle has its own rules, norms, and values. Some of these phyles are quite wealthy and have their own advanced technologies.
“Jennifer Government” by Max Barry: The novel is set in a dystopian future where corporations have replaced governments. There are affluent enclaves where company employees live, separated from the general population. The divide between the wealthy and the rest of society is stark.
“Snow Crash” by Neal Stephenson: While not exactly a refuge for the rich, this novel features the “Raft,” a massive floating city made up of scavenged materials, home to a diverse range of people including refugees and outcasts. It’s an example of an alternate society outside the mainstream.
“The Circle” by Dave Eggers: While not a gated haven, the novel explores a powerful tech company that seeks to create an all-encompassing online community where people live their lives. The company’s influence blurs the lines between private and public spaces.
“Super Sad True Love Story” by Gary Shteyngart: This satirical novel portrays a future where society is deeply divided between the technologically elite and those who are left behind. The wealthy elite live in protected enclaves with advanced technology, further isolating themselves from the rest of the population.
“Parable of the Sower” by Octavia E. Butler: While not centered around wealthy individuals, the novel explores the concept of intentional communities known as “acorn” communities, where people come together to create self-sufficient and sustainable communities in the face of societal collapse.
These novels offer various takes on the idea of exclusive enclaves, intentional communities, or alternative societies created by the wealthy or powerful. While the nature and motivations behind these settings might differ, they all contribute to the exploration of social dynamics and the impact of privilege and isolation.
Truth v fiction Sometimes you read things in sci-fi and just can’t think it could be real. But then sometimes sci-fi is really just a mirror to help us see what is happening around us in our own reality.
