Tanker traffic through the Strait of Hormuz has all but stopped since yesterday.
Read the @financialtimes.com Economics editor, Sam Fleming's analysis of what the impact the war in Iran will have on the global economy
www.ft.com/content/31bf...
Tanker traffic through the Strait of Hormuz has all but stopped since yesterday.
Read the @financialtimes.com Economics editor, Sam Fleming's analysis of what the impact the war in Iran will have on the global economy
www.ft.com/content/31bf...
screenshot of a Polymarket account (TJlaoyang) that shows they joined in Feb 2026, made one trade, buying 167,844.9 shares that the US would strike Iran on Feb 10, 2026 and lost about $2000 https://polymarket.com/@TJlaoyang
here's a random account whose only bet is (incorrectly) buying 167k shares that the US would strike Iran two weeks ago at 1.2Β’
that's the exact kind of thing that would look like an obvious signal in retrospect if they were right, but they weren't. how could you know the difference beforehand?
I don't think that's right. people start new accounts and place big bets all of the time. the things that makes the trades suspicious (the timing, the accuracy) are things that are only apparent in retrospect
ICYMI, FT Alphaville now has a Substack, which doesnβt cost a thing.
Yes, there are a LOT of newsletters these days. But if you appreciate stupidly detailed explorations of finance and economics, judicious shitposting and killer charts, then please sign up? ftav.substack.com
at some point in the next few years, I hope people eventually understand that the clunky code these LLMs write is:
- great for replacing common/low-complexity/low-risk programming tasks (a lot of those!)
- not so good for unusual/difficult/high-risk stuff (replacing legacy COBOL)
file this one away for a few months from now, lol
I've got the A1 too! but without the combo, so can't really do multi-color printing (which I think I need for lithophanes)
was looking at getting the AMS for it, but recently I've been considering just reselling my A1 and getting a P1S
A screenshot from blender of the frame base design
A screenshot from Bambu Studio of the two 3D printed sides of the frame base
The finished light-up map frame and base, showing the two relief maps of Rhode Island and of its Salt Ponds, backlit from the LEDs inside of the frame
then finally designed/printed this stand to allow the frame to sit upright and the power cord to run up the side frame to power the lights
A closeup of one of the corner spacers with a glued in bolt. The "wings" on two sides of the spacer slide into the grooves in the frame, which will then prevent the aluminum sheet from falling backwards out of the frame once it is secured on
A top view of one of the corner spacers with a glued in bolt with its "wings" slide into the frame's grooves
A top view of all four corner spacers laid out in the frame and slid into the frame's grooves. Once the aluminum sheet is set onto these, they will prevent it from moving forward or backward in the frame, while the aluminum sheet prevents the spacers from moving up/down or left/right, so their "wings" stay in the frame's grooves
A top view of the back of the frame after the aluminum sheet has been secured onto the spacers' bolts with four nuts on each corner
printed these little corner spacers to keep the aluminum sheet in place on the back of the. they have "wings" on two sides that slide into the groove of the frame and a glued-in bolt that threads through the aluminum sheet. worked out in a pretty nifty way that didn't require gluing the spacers
The LED lights behind the models on the inside of the frame. About a dozen LED strips are soldered together with short wires into rows, which sit on top of an aluminum plate which fits onto the back of the picture frame
The wires connect to a power supply and switch on the side of the frame
I glued the models onto a picture frame and replaced the back with an aluminum sheet that I stuck these LED strips onto. after a bunch of soldering, I had the lights set up and wired to a little switch on the side of the frame
Two 3D printed models of a Rhode Island relief map and a Rhode Island salt ponds relief map with a measuring tape next to them showing that they're about a foot tall when laid out next to each other
A 3D printed model of a Rhode Island relief map, backlit to show the elevation of the land/depth of the surrounding ocean
A 3D printed model of a Rhode Island salt ponds relief map, backlit to show the elevation of the land/depth of the surrounding ocean
I 3D printed the models in a matte white PLA, which made the elevations show up in this pretty cool way when backlit
A screenshot from blender of a 3D map of Rhode Island with an attached label and border
A screenshot from blender of a 3D map of Rhode Island's salt ponds with an attached label and border
A screenshot from Bambu Studio of two plates for printing with the 3D models of Rhode Island and its salt ponds
after that, I brought the files into Blender, where I added some labels, and a sea-level border on the models and then into BambuStudio to prep them for printing
A screenshot from QGIS of sixteen separate elevation tiles from the USGS in the Rhode Island area, showing elevation of land/ocean areas in black/white
A screenshot from QGIS showing elevation of land/ocean areas in Rhode Island black/white, stitched together from previous tiles
A screenshot from QGIS showing elevation of land/ocean areas in Rhode Island black/white, stitched together from previous tiles and cropped to the boundaries of Rhode Island
A screenshot of a 3D file, showing elevation of land/ocean areas in Rhode Island
first thing I did was grab topobathymetric tiles from the USGS, stitch them together, crop them to the state boundaries for Rhode Island, and then export an .stl file that I could edit for a print
A frame with two 3D printed relief maps of Rhode Island and the Rhode Island Salt Ponds, which are being backlit from within the frame
A closeup of the backlit Rhode Island/surrounding ocean relief map
A closeup of the backlit Salt Ponds relief map
wanted to share a little about a 3D printing/electronics project that I finished recently as a gift: this light-up relief map of Rhode Island and its Salt Ponds
I have some suspicions about why prediction markets are striking deals with newsrooms. www.theverge.com/business/881...
one of the world's largest smuggling networks has accidentally exposed itself because of a stupid e-mail configuration blunder, which has revealed around 50 apparently unconnected companies are all sharing back-office functions: www.ft.com/content/4310...
Mitigating factor: bro kind of sucked at insurrecting
in the context of whether "AI tools are replacing developers", we shouldnβt care that two CNBC reporters built some working version of an app that there are a thousand forkable versions of
there are some differences for sure. but at a basic level, there seem to be some misunderstandings over what the difficult parts of software development are and what we should be impressed by that are leading people to some wrong conclusions about what these AI tools can/are replacing
Paragraph from a story in The Atlantic: "But the labor market for office workers is beginning to shift. Americans with a bachelorβs degree account for a quarter of the unemployed, a record. High-school graduates are finding jobs quicker than college graduates, an unprecedented trend. Occupations susceptible to AI automation have seen sharp spikes in joblessness. Businesses really are shrinking payroll and cutting costs as they deploy AI. In recent weeks, Baker McKenzie, a white-shoe law firm, axed 700 employees, Salesforce sacked hundreds of workers, and the auditing firm KPMG negotiated lower fees with its own auditor. Two CNBC reporters with no engineering experience βvibe-codedβ a clone of Monday.comβs workflow-management platform in less than an hour. When they released their story, Monday.comβs stock tanked." The sentence, "Two CNBC reporters with no engineering experience βvibe-codedβ a clone of Monday.comβs workflow-management platform in less than an hour." is highlighted
also to be clear, I am kind of subtweeting this anecdote
I don't care at all about monday.com, but it seems very silly that their stock would tank over this "vibe coding", but not tank when I clone any of the 1000+ public repos here: github.com/topics/kanba...
what's the difference, practically?
tbh, it's less a "credit" thing and more the inability to discern what is impressive/novel while writing about how impressive the tools are
it's like being impressed that the "auto-painting machine" can make you Bob Ross paintings. you could've painted those yourself already!
part of why this is bugging me so much is that it seems many people are discovering the value of open source software, but crediting it to tools that are trained on it
these tools can do many amazing things, but I promise you that "barebones monday.com board that runs locally" was already possible
basic versions of these kinds of apps are easy to build (and the parts you can build in 15 minutes and run locally are not the tricky parts). that's why there are so many open-source versions on the internet, which is also a big part of why the LLMs are so good at making them
it only took you 15 minutes to vibe code a version of the monday.com kanban board? ok, I bet I can walk you through finding and forking a repo in less time than that and you still won't need to write any code. at some level, that's really not very different than what Claude is doing
something that I wish people would understand is that the kinds of apps that are easiest to "vibe code" a working version ofβkanbans, news aggregators, blogsβare also the kinds of tools that there are countless open source, free, forkable versions of on GitHub/elsewhere, and that's not a coincidence
A special FT magazine issue all about maps is out today, w great pieces by @okr.bsky.social and @theboysmithy.ft.com, plus maps by me on how glacial melt is redrawing Alpine borders, the battle to redraw America and more
Read it online here (though best enjoyed in print!) www.ft.com/content/efab...
I'm sorry the guy changing the rules on the fly is named what
as ever:
archive.org/donate?origi...
The Internet Archive is probably the org I've supported/urged ppl to fundraise for the most, after my own org, #MEAction.
Seriously, if you won't miss $5/month, it's always a good time to support the Archive.
Critics fear these tactics have a βchilling effectβ, discouraging people from filming. The result? βThese agents get to effectively be as unaccountable as any group in the history of our country,β says David Bier, of the Cato Institute.
Read the full story in todayβs #FTEdit π ft.trib.al/yAKlCgR
ICE agents are also invoking a federal criminal statute, 18 U.S.C. Β§ 111, which makes it a crime to interfere with federal officers while performing their duties. Civil rights experts say the statute is being misapplied, as recording agents does not constrain their ability to carry out operations.