Silicon Valley Redneck

Ian approacheth.

While I was out doing hasty chores, the sky was looking downright menacing in places, especially to the East.

Looking at the radar map?  Yeah, due East ±80° or thereabouts is that big storm, and it does appear to be drifting this way.  Hot cocoa mix on standby.

Puzzle maybe solved; flashback to an old project

The mystery of the cursed high-speed UART may (I emphasize may) be sorted.  And I'm having a flashback to the mid-1990s.

In today's effort, I eliminated several more things, and basically narrowed it down to: if either receive interrupt happens (that'd be character received, or character(s) received followed by a gap), there's some probability of things going wonko.

What it finally came down to was: this UART, unlike its predecessors, is running on a clock generated by a Spartan 6 DCM, and thus is more or less unrelated to all the other clocks in the FPGA.  ("How is this clock unlike all other clocks?"  Wait, that's out of season, innit?)

Anyway, synchronizing all the interrupt requests to the CPU clock seems to have solved the problem, or at least made it vastly less frequent.  Still gonna run the test for a while before declaring victory and checking in the updated code.

And the flashback?  Back in the day, when Signetics was still Signetics and not Lowe's, someone there had a brilliant idea: a sort of programmable DMA controller chip, called the I/O Processor or IOP, whose part number escapes me at the moment though I think I still have at least one of the chips around.  They made an initial batch, and I got to design a demo board for it.  (I was working for IPT back then, and one of IPT's activities was developing demo boards and software for new chips; that was a seriously fun part of the job.)

Well, it was awesome.  Right up until it started flaking out every so often.  After much diagnosis, involving a logic analyzer hanging on the address bus, I came to the conclusion that the request inputs were not being properly synchronized, so sometimes a race condition would lead to it taking the wrong vector or even loading part of the address from one vector and part from another.

Alas, they didn't respin the chip to resolve the problem; they just killed it off.

My problem, being in FPGA logic, just required a respin of the code, assuming I've correctly identified and solved the problem.  So it doesn't kill the project.

(Checks back on the test: still blinky-blinky, and no evidence of trouble.)

And so, as the month comes to an end, it seems I can include good tidings in my report.

Update:  Code checked in.  End-of-month paperwork shuffled.  Back lawn mowed.  Straw in chicken coop changed.  Hatches battened for Rainy Weekend Ian.  Test is still running happily, so I'll assume the problem is fixed and proceed (in the coming days) with tidying up and finishing stuff so far left incomplete, plus getting the RP2040-based serial dongle working.

Time to fix dinner now.

[Saber-rattling intensifies]

There's much noise, from both sides of the Tinfoil Curtain, regarding that explosion at Nordstrom....

... I'm sorry, I'll read that again: the explosions on the Nord Stream pipelines.  Both pipelines, three pipes in total, ruptured while not actively carrying gas.  Definitely not going to be carrying Russian gas to Europe for the winter, you betcha.

Who could have done such a dastardly thing?  Means, motive, and opportunity eliminate some players, but leave a heck of a lot, including freelance environmental radicals with some financial backing and a little expertise supplied.

NATO says it was the Russians.  Putin says it was the West, and something Biden said a while back suggests American sabotage.

Lawdog has another theory.  The complications of dealing with methane at high pressure and low temperature in the presence of moisture, plus famous Russian maintenance standards.  Natural hazards plus stupidity and incompetence?  Naw, that's crazy talk.  Besides, that kinda stuff doesn't sell emergency measures.  We must have emergency measures!  We must have global thermonuclear war!

Geopolitics has no room for people who offer calm, rational, science- and engineering-based explanations for events.

The coming AI Apocalypse: not what you think

Oh, noes!  When the computers get smarter than us, they'll decide humans are superfluous and eradicate us!  Maybe they'll keep a few of us in zoos, for robots to come look at!

Well, it makes a good movie.  Or a bad one, as the case may be.  But...

OK, this kinda congealed as I was commenting on a post at BRM.

Part of the background to a sci-fi epic I'm not writing is this set of assumptions:

• Current efforts in AI will not lead to true artificial intelligence; the underlying methods are a dead end.
• The various sorts of pseudo-"intelligence", while not rising to the true intelligence even of a puppy, will nonetheless become better at most skilled trades than all but the top humans in those trades.
• With skilled trades (medicine, law, aviation, and so on) practiced at above-average level by machines, there will be no opportunity for humans to practice those trades and attain the top levels, where the machines can't reach.

There are various other assumptions, some of which I'd need to discuss with a rabbi if I were actually going to write the dang novel.  But those are the key ones.  (The others mainly relate to true AI, developed some centuries in the future.  One is somewhat borrowed from Red Dwarf*.)

Now consider advanced autopilots.  Human pilots are still legally required in the cockpit, and will be for the foreseeable future, but we can pretty well see a future in which all routine operation of commercial aircraft is handled by the computers.  And why not?  The computers will soon be capable of flying the plane better than most human pilots.  And when a situation arises that's beyond the computer's competence?  You hope the human in command is Sully, but it's more likely to be the crew of AF447.  And, with routine operation automated, the only real experience the humans will be getting will be simulator time, while cockpit time will lull them into state of complacency.  Eventually, all the top pilots will have retired, and there'll be little point retaining the ones who are no more capable than the machines.

OK, let's extend this to skilled trades in general.  How many specialists can you think of who could perfectly well be replaced by an expert system with some combination of natural-language communication, vision, and manipulators?  I mean, for routine stuff, not for those special "there are only three people in the world who can do this, so get on the waiting list and hope one of them will get to you within your lifetime" matters.

Doctors, lawyers, farmers, carpenters?

Extrapolate a generation into the future.  All the humans who used to practice these trades have long since retired, and most have died.  The trades have morphed into automation-oriented forms that no one comprehends.  And, of course, all the machines that do this stuff are controlled by cartels, with the possible exception of some outdated ones that have been salvaged and are being operated illegally outside the system.

And now the next Carrington Event comes along.  Or a flurry of cyberattacks, or EMP.  We're back in the Stone Age, minus the skills that enabled everyday humans to survive in the Stone Age.

It's an AI Apocalypse, only without any hint of, nor need for, malevolence on the part of the AI.

 

* Which didn't take it back far enough, nor elaborate on the reason.  There was one episode in which the ship's computer had its intelligence boosted beyond recognition, which resulted in its life expectancy dropping precipitously.  OK, so maybe there's some mysterious reason.  Or... just a thought, just now... consider a ship's computer with an IQ of 6000, which is what that computer was supposed to have at the beginning of the show.  What's it gonna do with all that intelligence?  Routine operations & maintenance don't require IQ points, just lots of little specialized computers.  That leaves the intelligent computer pondering the imponderable, unscrewing the inscrutable, and descending into boredom and madness.  Gotta be a general principle here, involving level of intelligence, amount of input available, and ability to manipulate the world at large.

 

Afterthoughts:

Methinks a lot of the trouble we're seeing in the West is the result of intelligence having no real challenges to overcome.  There being no existential threats in everyday life, people need to manufacture them.  See the result: War, Famine, Pestilence, Death, and Madness are on the rise.

... Which ties into some further background of that novel I'm not writing.  There's an aspect I suspect wouldn't work: a planet kinda-sorta like Israel, only not surrounded by enemies.  Would it be like Israel, only prospering in peace?  Or are the enemies necessary?  Put half a dozen Jews in a room for an evening.  Now extrapolate to planetary scale, and a few centuries.  (This probably works for any other tribe of humans, and for most other mammals for that matter.)

Morning sidetrackery

Before I get back to the toaster project....

Someone had mentioned the ESP32-S3 as an RP2040 alternative.  Slightly pricier, but faster CPU cores, available with a boatload of RAM, and has the WiFi / Bluetooth hardware if you want it.

Hm.

Yes, it looks plausible for some purposes... including some little solar-powered sensor platforms.  Sleep-mode current draw ain't wonderful, and I sure wouldn't use it in anything powered by a primary battery, but for a solar-powered dingus with a ~400mAH LiFePO4 cell to cover the dark times?  Yeah, could work.

I actually have some ESP32-C3 modules sitting around waiting for me to find time and attention span to evaluate the BLE capabilities, in particular the range attainable with reasonable antennae.

And: some of the parts are rated for the full industrial temperature range.  Some, though, are only rated from -40 to +65°C.  I don't really expect that to be an issue for anything I'd be using around here, but were I developing a product for sale I'd have to pay attention to such details.

For most of what I have planned, either an RP2040 or something with seriously low sleep-mode power seems a better match.  (C'mon, Atmel/Microchip: 18-month lead times are not reasonable!)  I'll keep the ESP32 in mind as a general-purpose MCU*, though.

 

* I think Marvell should come out with a line of motion-control chips and call that product line the Marvell Kinematic Universe.

Just what I needed: another puzzle.

Trying to finish up this toaster project, and:

Turns out the toaster reboots randomly when communication is running.

I assume that something to do with comms, either the UART driver code or the routine that processes requests, is somehow getting hung up and allowing the watchdog timer to do its thing.

But... I remove pretty much everything, a piece at a time, and as long as the receive interrupt is happening (and the data being read, though I haven't investigated what happens if it isn't read - probably the interrupt just happens continuously) the reboot happens every so often.

Occasional gibberish before the reboot suggests stack corruption; it looks like some random code is getting executed out of place.  Maybe.

The stack seems to be consistently in the right place in the interrupt routine.

Grrr... this is an AVR core synthesized in an FPGA along with a bunch of application-specific stuff.  Funsies.  But most of it has been working fine since (checks records) spring of 2012 or thereabouts.

There's a filksong that seems relevant here... I think it's by Duane Elms... aha!  Hard to find by name on the 'Net, but it's on the Bayfilk Cyberlite tape.  "Homicidal", indeed by Duane Elms.

Now the problem's not the mem'ry, and the problem's not the plug,
And the problem's not the program (though it has this little bug).
After Service replaced everything, with not one module missed,
They said the damn thing's haunted and they called the exorcist.

(Hey, that tape also has "The Roving Repairman-O"!)

Update:  Well... lots and lots of "that wasn't the problem" and a complete lack of "well, there it is!"

It's... almost like this project has a sort of digital later-strain SARS-CoV-2 infestation, with the cause obscure and all manner of inconsistent symptoms.  The one thing I did that seemed to make it work involved (1) completely disabling the FPGA-emergency-reset logic, and (2) eliminating one term from the CPU reset logic.  Which might have suggested that the term in question was having noise coupled in, elsewhere on the board, from the receive data line, except for the minor detail that the term was actually nailed to zero in the FPGA logic and therefore should have been optimized out anyway.

When it does reboot (which sometimes happens even with the FPGA-emergency-reset logic disabled), there's sometimes a burst of gibberish, and sometimes some normal(-ish) output that ought not to have been happening, and sometimes (well, exactly once) some debug output that was meant to catch what I originally thought was the problem.

Anyway, either there's some serious weirdness with the new comm code (whether the MCU or FPGA side of it) or some code that's working fine for the last decade has a lurking problem that's just manifested itself.

Hmmm..... I wonder what would happen if I ran this on an older board.  The actual control functions wouldn't work, what with the older board being a different model, but the comms should.  It's just vaguely possible that this year's (pure unobtainium) FPGA chips are slightly flaky, despite supposedly having come through proper channels with documented provenance.  (Or, maybe not.  I think ToasterCo has successfully run the code for an older model on the same board on which this problem was spotted.  But I'm not entirely sure.)

Epic storm is dissipating?

Latest forecast for the weekend is for light rain.  I knew there was a reason for having distance and/or mountains between us and the ocean!

And, given the reports coming out of Florida?  I'm real glad I'm not there.

As usual, the news focus is on the most dramatic spots - or at least the most dramatic spots accessible to TV crews - so information on places outside the focus is sparse.  Maybe I should check on ex-GF in Titusville; from what I can find on line, that area is getting a fair bit of wind, rain, and flooding, but nothing like what Fort Myers saw yesterday.

Gettin' meta

So the UPS truck came...

... bearing a box...

... containing two UPSes.

Another note on the garden thing

Yes, it's supposedly a well-established fact that it's possible to feed a family with a quarter-acre farm and primitive methods.

But.

That involves some assumptions:

• The land has been prepared for agricultural use for some years - most likely has been used as family farm land since time immemorial.
• You know what you're doing, having grown up doing quarter-acre subsistence farming.
•     ... In the same region and climate.
• The family includes a few reasonably healthy kids who can be put to work.
• Nobody commutes to a day job, or goes to school, or any of that stuff; everyone has time for working the garden.
• Outside expenses are minimal: no paying for water, electricity, health care, telecoms, and suchlike.
• The neighbors are doing the same thing, and won't complain about your pig.
• You understand the practicalities of food storage, so your winter food supply won't rot.

I'm gardening mainly as a hobby, but also to make the mistakes (and prepare the land) while it is still mainly a hobby.  Those with quaint notions of returning to a simpler way of life, who may suddenly be trying to support themselves off the land for real?  Well, when romantic ideas meet with unfortunate reality, reality wins.  A packet of "survival seeds", a trowel, and a few hundred square feet of lawn won't get you very far.

Garden exercise

Been digging up sweet potatoes.

The sweet potato patch, like so many other parts of the garden, has yielded more learning than actual food.

Main lesson: the soil should be much looser.  Add sand and compost, and subtract weeds.  Especially subtract Johnson grass, but weeds in general (including all manner of grasses) make delicate digging a challenge.

Oh, well.  I did get a few pounds of probably-usable roots, plus lots of tiny roots and sprouts that are now ensconced in a tray of potting soil, water, and rooting fertilizer in the basement.  And I got some exercise.  Still need to dig up the onion zone and see how much I can salvage.

Meanwhile, the forecast for a hurricanian weekend is firming up.  Guess I'll be having a couple of very indoors days,

And just what are the Globalists up to?

Bill the Gates, quoted here:

“I admit that political polarization may bring it all to an end, we’re going to have a hung election and a civil war,” he said.

Hm.  Definitely a hung election, and definitely a war.

Well... various prominent Democrats have, the past few months, already been spreading the "stolen election" narrative* ahead of the midterm elections.  So they're planning shenanigans, and still expecting to lose.  Expect riots.

(I'll note here that the text-message spam addressed to my father, on a number the Democrats are absolutely, positively not authorized to use, now has implied threats mixed in with the usual scare tactics.  "We've given you 17 chances already to support us!  You're now on the doesn't-support-President-Biden list!  Are you sure you don't want to reconsider?")

And the war?

Well, the Globalist agenda does call for reducing the world population by 90% or so.  Wars everywhere could accomplish that, along with the destruction of the infrastructure that supports the lifestyles of the rich and famous unsung working class.  Presumably the rulers already have their volcanoes hollowed out.

Song:

 

* They've been using it for decades, most conspicuously after 2000.  The one time a Republican (of sorts) borrowed it from them, it was Sedition and Insurrection.  Kinda like how they get all bent out of shape over "Republican dark money networks" when the Progressive dark money network was well established when I first learned of it in the early 1990s.  Or demand that we get corporate money out of politics, then push ESG rules to compel corporations to support their political causes.

My, how misunderstood terminology can cause problems...

... Many decades later.

Just looking at the specs for the UARTs in the RP2040, implemented using an ARM standard library component.

There are hardware RTS and CTS signals.  Hey, can I use RTS to control an RS-485 driver?

Dig into the specs.

Nope.  And the reason goes back to misunderstandings, misapplication of signals, and, way back when, a standard that was designed for asymmetric application.

Let's all put on our Wayback hats and vortex goggles....

The labels RTS and CTS come from the RS-232 standard.  You, know, those old-timey DB-9 ports on older PCs....

... Wait.  This has already gone doubly wrong.  The standard used a DB-25 connector, and the 9-pin D connector is an E-size shell, not B.  IBM, in its infinite wisdom, used a DB-25 connector for the Centronics-style parallel printer interface, instead of the AMP Blue Ribbon connector found on the actual printers, a male DE-9 connector for the serial ports, instead of the standard DB-25, and a female DE-9 for the video output.

(IBM got the size wrong, but the gender right: PCs are DTE, so the male connector is correct.  Many other vendors, by that time, had settled on the "female connector on equipment; male connector on cable" convention, regardless of the nature of the equipment.)

Anyway, the actual RS-232 standard had various handshake lines, including (among many others filling up most of those 25 pins) RTS (Request To Send) and CTS (Clear To Send).  Actual implementations typically have some subset of these, usually including RTS and CTS.

But.

The standard assumes two types of equipment: DTE (Data Terminal Equipment, such as Teletypes and computers) and DCE (Data Communication Equipment, typically modems).

Almost immediately, people started cutting out the middlemodem and connecting DTE (terminals) directly to DTE (computers) with local cables.

Now, properly speaking, when one connects two RS-232 devices, it's supposed to involve a straight-through cable - no wire swaps.  RTS (DTE) goes to RTS (DCE), and CTS (DTE) goes to CTS (DCE).

When the modem is taken out of the equation, things need to get crossed over.  Swapping the transmit and receive data lines works as expected.  But... if you need hardware flow control?

And so people started connecting RTS to CTS, CTS to RTS, and redefining RTS from Request To Send to Ready (for the other party) To Send - i.e., what CTS coming out of a modem means.

Then this (stapler) redefinition of RTS ended up hard-coded in generation after generation of UARTs, with no proper RTS function available to control a line driver.

So now, when I see that there's hardware RTS support, and I look up the function to confirm that it'll do what I want, I'm invariably disappointed.

Oh, well.  For the task at hand, I can just leave the transmitter enabled.  If at some time in the future I need to support half-duplex operation, it's a simple matter of digging into the doggone UART driver code and adding a couple of lines to set a GPIO pin when starting transmission and clear it when transmission is entirely complete.  Which... a lot of UARTs don't provide transmission-complete status, or at least the ability to interrupt on that status.  I don't recall whether the ARM UART is one of these.  But that's another rant, which I probably already published back in 2012 or something.

Sans-serif fonts, wafflewords, perennial confusion

I keep seeing references to some sort of new local network.  HurricaneLAN, it's called, or something like that.

Or, rather, WaffleWordHurricanelan.  Except that apparently it's meant to be WaffleWordHurricaneIan.

Let's try that in a serif font:

Or, rather, WaffleWordHurricanelan.  Except that apparently it's meant to be WaffleWordHurricaneIan.

Yeah... The ubiquity of sens-serif fonts, especially with aggressive kerning (or keming, as the case may be), makes for confusion.  The fashion of smooshing a bunch of words together, optionally with CamelCase, and hanging a waffle sign off the front doesn't exactly help.

(It does appear that Hurricane Ian may be paying us a visit over the weekend.  He should be pretty well tuckered out by then, after whoopin' it up in Florida and than partying all the way up I-75.  Do we need to fix up the guest room for a partied-out, hung-over tropical storm, maybe with a case of tropical depression?)

Some people just live inside their own heads.

Yeah, yeah... an astounding revelation, innit?

The latest silliness: Republican / Conservative / Christian men are planning to impose a dress code for women!  ZOMG, red robes and white visors!

Er.

Ever see how regular gals dress in the Bible Belt during the warm season*?  Iran, it ain't.

Speaking of which, how come the same people who push this nonsense always seem to be totally fine with an Administration that keeps propping up the Iranian regime, which does enforce a strict dress code for women?

* Summer fashion in the Deep South: a bikini and an AR-15.  If a woman is seen in public without an AR-15, the fashion police will issue her one.

A moment of weather

Well, several minutes.  But still: today's weather was very concentrated indeed.

The early forecast showed scattered thunderstorms and a fair chance of a little rain, but all the thunderstorms swerved around us.

Around dinnertime, the sky got menacing.  Dark clouds, thunder in the distance, and moderate gusty wind.  I hastened to finish grilling, get back indoors, and serve dinner.

Not long afterward: GOOOOSH!  Sudden high wind and heavy rain.  Stuff from the deck got blown a considerable distance.

Then it stopped.  Rain gauge shows 0.57", all of it in those few minutes.

... Maybe my EDC needs to include a raincoat and a can of lightning repellent?  And an anchor, lest I be carried away to the Land of Oz?

Example code with automated builds is just so helpful!

So, finally getting around to trying to Do Stuff with the Raspberry Pi Pico.

Got the C/C++ SDK installed OK.

By following the instructions, I can build the blink app, copy it over to the board, and get the LED a-blinkin'.  Yay!

OK, so next....

Next is trying out the tinyusb_dev_cdc_dual_ports app.  Which... isn't in the source tree, but the cmake operation pulled it into the build tree, somehow.  Good?

It builds, but, well.  I can't tell where it's finding the actual main.c, which makes the example rather less than useful when the idea was to get a quick start on my own project by copying the example to a new directory and editing the source files.

The Makefile was somehow created by cmake, which is not something I've gotten around to studying.  It looks like there's a src directory, but either it doesn't really exist (perhaps having been created and populated on the fly and then cleaned up when no longer needed) or it lives somewhere nonobvious.

Oy.  Two levels of build automation to obfuscate basic stuff that one would think an example project should exemplify.

Also, while there's at last somewhat useful documentation of how to use the UARTs with the SDK, I'm not finding much on how to use the CDC device stack, and the example provided doesn't explain much.  This should be easy: all I'm trying to do is bridge the CDC device (/dev/ttyACM0 or whatever on a Linux host into which the board will be plugged) to one of the UARTs, bidirectionally.  Having control pins for driver / receiver enable would be nice, but I'm not expecting that.  But, much learning ahead, and precious little of it relating to the project at hand.

Update:  Maybe the example I'm looking for is the "hello world" app... which comes in two flavors, UART and CDC-ACM.  But!  The source code is identical.  The difference is mysteriously incorporated by cmake, somehow.  At least this does point me to the fact that stdio can be configured to use either a UART or CDC.  Now I just gotta figure out how to write my own program and get it to link with the doggone SDK....

And I thought the years-ago SDK/IDE for the Kinetis MCUs was confusing.  At least I could see where the source code was!

Bit of a workout yesterday.

I went over a small patch of moderately large weeds with the new intermediate-sized rototiller, partly to deal with the weeds and mostly to try out the new apparatus.

It's cumbersome.  It can be pushed along with the wheels disengaged, but going uphill that way (or even just across grass) is a pain; something about the machine weighing 170-ish pounds.  Propelling itself, it's a slow and cumbersome beast, moving along at tilling speed - and it's necessary to hold the handles aloft, to keep the tines off the lawn; there's no "drive but don't till" mode, nor rear wheels.

In actual use, it does a good enough job on a previously tilled but neglected area.  Chaws up the weeds pretty good, though the taller / tougher ones tend to get wrapped around the tilling axle.  Does a proper job on the soil.  The drag, though, just sinks right in, so setting the tilling depth is accomplished by (yes, you guessed it) holding the handles up*.

Corners are a pain.  It's not very maneuverable.

All in all, I think it'll be well suited for the intended purpose, i.e., re-tilling long but somewhat narrow areas between rows in the Victory Garden.  Add to notes: arrange next year's garden to facilitate running the new tiller though along fairly long corridors.

For tight spaces with small weeds, the mini-tiller is the thing.  For spring and fall tilling of an entire field, the big one gits 'er done quick.  This one was meant to fill the gap between them, and it seems like it should do that, though getting it to and from the area of use may be a nuisance.  (Another reason to put a medium-small shed down in the garden area.)

* I keep wanting to attach some sort of rear plate between the skids on the big tiller, so it'd kinda-sorta level out the soil behind it, as well as having something a little closer to actual depth control.  The skids just sink in, so, if I really want to control the tilling depth, I use the height setting on the 3-point.  Welding a crosswise bar to the bottom of the drag on the new tiller might make sense... maybe.

XKCD nearly exaggerates.

A couple of episodes back:

But!  You don't even need to collect and lug around all the EDC products on the Internet.  Just collect a few people's recommendations, merge them as sensibly as you can, and then examine the situation.

You'll find yourself with a backpack-sized first-aid kit as a bare minimum, two pistols plus at least two spare mags for each, at least two knives and one flashlight, bear spray, a sewing kit, a shoe-repair kit, your cellphone, a burner phone, sunglasses, NARCAN, a ham radio, a lock-picking kit, a set of wrenches, a change of clothes, and probably a 2-day supply of food and water.  All this just for basic daily existence.  Additional minimum requirements for investigating a bump in the night: level IV body armor, a specially-configured rifle and at least 300 rounds of ammunition, night vision goggles, and a few flashbang grenades.

I limit my EDC to stuff I'm somewhat likely to need, that'll fit in pocketses.  Keys, flashlight, dinky Swiss Army knife (featuring scissors and nail file), actual pocketknife, personalized mini first-aid kit (basic cuts & scrapes stuff plus Benadryl and epinephrine), and cellphone (in a case which also holds a spare house key, a couple of credit cards, and a grocery-store loyalty card)*.  If I'm going somewhere, add a wallet and a car key.  Might be carrying various other items, depending on the location and circumstances.

If I'm not out in the actual wilderness, I don't need to be carrying a wilderness first-aid kit.  I'm not in the habit of visiting war zones, nor even crowded places.  Excessive encumbrances make daily life unnecessarily awkward.  I don't need to be a walking ambulance (and I don't have the training to use those advanced first-aid goodies, nor the spare time to get trained and keep current), and I find that maintaining some reasonable degree of situational awareness and not going stupid places with stupid people to do stupid things at stupid times will generally keep me clear of mini-wars.

* This is somewhat redundant, but if I'm in my pajamas I only have one pocket, so the phone rides around in that and thus I have an emergency house key, just in case.  Yes, I'm paranoid about doors ending up locked behind me - ever since the second time I locked myself out of my apartment back in college.

Out of phase

Well!

In a normal year, or even 2020 or 2021, my metabolism would have shifted into Spring weight-loss mode sometime around, well, Spring.  I would have been at Summer weight for the hot season, instead of carrying around excess polar-bear insulation.

That metabolic crud I've been having this week?  I've suddenly lost a couple of pounds, and my pants are decidedly loose this morning.

... Just after I bought three new pairs of cargo shorts in my current size, because I clearly wasn't about to get back to the one-size-smaller (whereof I believe I still have a goodly stash).

Yes, this does feel like the shift to weight-loss mode.  I may need to break out some long pants this Winter - fancy that!

Gee, I wonder what they want.

Daily e-mail about inbound postal mail shows something from "Land Insider USA".

(Imagines holding envelope to forehead.)

No.  The answer is no.

(Imagines opening envelope.)

Would you like to sell your recently-acquired land for an insultingly low price?

Funny thing: the return address is in Nashville, but it's postmarked Sacramento.

I suppose I'll open it when it gets here, and see what they really want.  Mind you, I can't, on my own, sell any of our real-estate holdings, but I think these people don't bother looking at the actual ownership records.  Heck, somebody wanted to buy our home from my father, and his name isn't on the title to this parcel at all!

Update:  Yup, insultingly low offer for one of the Northern Territories.  Guys?  Firstly, you need all three of us to agree to sell it, and secondly, we've owned it for just over a year and thus aren't likely to be clueless about what it's worth.  Both of these facts are public record... as well as what we paid for it, not so long ago.