MilStdJunkie

To explain how that relates to your point re: stages. The stage of our perception isn't stable. When we started voyaging on ships, that changed it. We went to orbit, whew boy. Today, maybe our terminally-online presence is changing our notions of distance and "here-ness". These sorts of changes can often be a trigger when it comes to these perceptual shifts, as people grow up with a somewhat larger universe than their great grandsires.

Some of the most beautiful moments in science - for me, anyway - come about when there's a refresh on perspective. You'll have something like the ptolemaic system, or phlogiston, or the luminiferous ether, and a tiny push in perception reveals a bigger picture, a deeper understanding.

That push is often subtle, but incontrovertible.

Modern cosmology really feels like it's on the cusp of something like that today - perhaps in how spatial dimensions emerge, or some other thing - underlying our more stolid mental models.

No, I apparently dodged the trap of words pretty handily there when I meant to type "subscribe".

I apologize, I'm dealing exclusively in XSLT today and I can feel my brain screaming.

Not quite. I don't ascribe completely to the notion that we can't figure stuff out. But we do need to be careful about assuming that things like 3d topology (aka "inside", "outside", when the mathematical spaces we're looking might have no equivalent concept), cause/effect -- or even the spatial dimensions themselves -- are fundamental. Particularly when we're talking about black holes. It's hard, because our brains are optimized to deal with this very specific slice of spacetime we've evolved in, so we keep falling back into the trap of our words.

When I hear debates about whether or not the universes are black holes nested inside each other, or what the fate of the cosmos is, I often have an instinctive reaction to question whether or not concepts like "inside", "nested", or "fate" have actual fundamental qualities at the scales and timeframes under discussion.

Everything about the official inflation calculation is almost totally borked in terms of generating useful information. Like, information that would be useful for someone doing something real, like future historians or Zombie FDR.

There's probably dozens of mechanisms at work to explain "why??!!", but the root cause is - like so many other damn things in America - financialization, from its birth post-1972 with vehicles like EUROBOND, to its "pedal to the metal" moment with all the liquidated commie national assets, to the craptacular commodity runup crisis courtesy of Bush II wrecking the Levant and West Asia. Oh, right, and the financial system ending in 2008. When we seamlessly blended up - legally - private with public risk. Oh yeah. Sixteen years ago.

Anyway, financialization. We don't account for money supply accurately, they're dynamic with price indices now, blabbity blabbity blah blah, 1962 dollars were probably worth waaaayyyyyyyyy more than we think they were.

Truth. Splitting labor by race, sex, and whatever other box you can draw.

A trick as American as apple pie and, well, hiring weirdos to screw with unions.

Something fun to think about: the explosion in preposterously complicated documentation systems mirrors the degradation of union representation in skilled trades[0]. That's when capital should maybe stop and think about what, exactly, a union really is: bolshevik, or guild?

If a union cost you fifty million dollars, but your new slam bang doc system is costing you three hundred million, by itself, and your production times are 10x worse[1], it might be time to go back and re-examine your assumptions about the world. Because - surprise! - it's not bolshevism, it's f#&$ing compliance*.

[0] You're going to be tempted to say computing is sole cause here, but that's not quite correct. Scientific Management tried to do the same thing for US shipyards in the later 1890s, but it largely failed, because papered (skilled) labor knew what it was doing.

[1] Not to mention there's some things - complex welds, to take one example - that you are literally incapable of doing anymore.

Wow, "pay as you go html to markdown" seems like a very hard sell in this year of our lord 2024. I've got a brace of converters for various parts of this problem, some of them are pushing a decade in age. You have products like Zapier or the open source ActivePieces to restrain extraction to specific elements. Finally, several pre-wrapped AI models for this specific purpose.

Speaking for myself, as a layman, I had always been walking around thinking of hadrons as balls with other balls inside, with forces mediated by tiny/insubstantial yet-to-be-discovered balls.

Yes, I knew they weren't balls, exactly, but still. Newer models with greater sensitivity have sort of pulled a fast one: that maybe the balls are a side effect of the way we look at them. More precisely, to look at them, we need to smack them hella hard, to make the balls come out where we can see them. But the smacking might be a part of why the balls look like balls!

The native hadron in its natural habitat might be something fantastically more complex, a sort of cell biology of energies all competing and vibrating, and they're part of the interrelated forces of the nucleus itself - we might even use that obsolete term return, the "nucleon", to represent this complex. Jiving against all of this are new ideas about what mass actually is, defining spatial attributes as degrees of freedom, and all sorts of new thoughts. It's super exciting.

So that's what I mean by "shocked how little we know", because what we did think we knew got oversold as a World of Balls.

But who knows?! Maybe this neat laser thing can help find more answers.

Nice to have an alternative to Lilypond (lilypond.org), but given the extreme complexity of notation, I would bet that any brevity gains are short lived.

For those Asciidoc freaks among you, Lilypond is fairly easy to get running in your Asciidoc toolchain of choice. I use the DocBook PDF pipeline, and the lilypond output is quite nice looking. It's awfully TeX-like.

When you stop and look at QCD in the big picture, it's sort of shocking how little we know - like, really, really know - about the internal structure of the proton, or even the nucleon!

It's the curse of "probing" with massive energies. No one's a hundred percent certain of whether they're detecting something that's actually there - like there there - or whether they're looking at by-product of enormous collision energies.

Physicists are smart people! I could never do what they do. But there's a limit to certainty, and inside the proton especially there's unknown first principles at work. Bringing the precision of photons and lasers into this nucleon party is going to be huge. I can't wait!

I curious how good the infrastructure -- the plumbing/electrical/network/appliance hookups -- are in this sort of thing, and how repairable the fixtures are. If I have to sledgehammer a wall out to replace a bad outlet, I think that would get pretty old pretty quick. Having built some boats, and having lived in an old abandoned boat through college (somehow getting the thing shipshape on undergrad money, although I took shortcuts that would curl my hair now) I can testify that the hull is the cheapest part of the boat. All the other stuff is the hard part. Houses are sort of like that, but a lot less critical than a boat hull at sea. The walls and roof are just chunks of stuff.

Life isn't an RPG where you read the gamefaqs and do an ULTIMATE BUILD. Your sense of self is a story, it's a narrative we tell ourselves, something we've evolved so that we can simulate our environment over and over and over and over in our heads without losing the thread. The idea that you're a meat machine being steered through the world by some disembodied entity - that you happen to identify with - is an illusion.

I'm going to feel like a real tool for quoting this, but I like the line they added to the new Dune movie: "The mystery of life isn't a problem to solve, but a reality to experience."

Now, this could be due to the fact that I've hed singular, terrible moments of energetically bad luck. Lots of them. And this sci-fi pseudo-dao is just a part of my personal narrative that keeps me from racking a slide on just the one bullet. But everyone, everyone everyone is going to have one of these moments. Dao isn't the solution, but resilience is part of one.

Palm was incredible. Enyo, webOS, the list is a long one. It's hard to overstate Palm's effect on handhelds, how they successfully challenged Blackberry to the benefit of both companies, and the tragedy of its demise[1]. Jon Rubinstein, Palm's creator, is bitter about its fate to this day.

I hiked the entire Appalachian Trail with my Palm TX, dutifully logging my thoughts and ideas on the two thousand mile walk from Georgia (USA) to Maine, saving them to a tiny SSD, then beaming them all to a journal when I hit a wifi spot in a town library or somewhere.

I still think back on that thing, and the absolutely AWESOME folding keyboard I had for it (2 oz, smaller than a pack of cigarettes folded up, folded out into a rigid keyboard that had a tray for the Palm, really useful typing on your elbows on the floor of a tent). It was a full commodity computing environment: calculator, cached wikipedia, word processor, email, pdf maps/guidebooks/almanacs. Honestly, if you added Maps to that thing, I would still use it today, no question.

Palm was - once Elevation Partners sucked all of the cash from the place - hoovered into the giant value-destroying machine of HP, where it promptly died, as do all things that enter HP post-HPUX.

[1] The moral of which: "Do Not Trust Private Equity. Ever"

Depends on what you mean by "machine gun"[1]. By the most common military definition, no. The recoil force would probably throw it to the ground, and even if it did not, the accuracy would be so laughable that it'd be more efficient to just load cyberdog with the same weight in high explosive fragmentation.

A SMG or autopistol . . maybe. Something like the Skorpion, firing .32 ACP, might be low recoil enough for aimed fire.

[1] There's a few categories here, and popular/legal categories differ from military ones. Legal "machine guns" include firearm capable of automatic fire, but this includes: 1) Autopistols and submachine guns (SMGs) fire pistol calibers, from .22LR (American 180) to .45 ACP (Thompson); 2) Assault Rifles, firing intermediate power cartridges like .223 and 7.62x39mm; 3) Battle Rifles are firing full power rifle cartridges - .308 Winchester and 7.62x54mm- from a box or drum magazine and can fire in full auto[a]; 4) General Purpose Machine Guns (GPMGs) capable of sustained belt-fed rapid fire with full-power rifle cartridges; 5) Here we're entering the realm of support weapons[b], firing Heavy Machine Gun rounds like .50 BMG, 12.7mm, 14mm, and other weapons like the Mark 19 automatic grenade launcher or the M134 Minigun (which fires .308 but a whole lot of them).

[1.a] Lots and LOTS of Internet Tough Guys will insist that the M14 or FAL on full auto is totally controllable by Real Men - i.e., it's only the sissification of our cucked armed forces that have brought on the curse of the Intermediate Cartridge. I'll leave judgement to the reader. Incidentally, the Venn Diagram of "Battle Rifle" and the older 1900s-1930s term "Light Machine Gun" is, functionally, a single large circle - it's just the advancement in materials and machining that make them any different. So why didn't LMGs take off back when men were Manly Men and could handle the POWER? Heh heh heh heh yeah. Physics wins again.

[1.b] Needless to say, a whole other category of internet moron insists that .50 BMG on autofire is controllable by a "man enough" infantryman operating singly. I'm sorry that this opinion exists, like, as a human being, but make your own judgement, as always.

Yeah, my bad, chucking rockets isn't what I had in my head, but I always try to use "chuck" as a verb as often as possible. What's actually the critical word is "invasion": no one[1] tries to take your land when you can hawk[0] a nuke at theirs. If you can give me a counterexample, I would be super excited to hear it.

[0] In your FACE, "chuck"! There's plenty of verbs in the sea.

[1] This is a fun one. "No One" would be "another state", someone that has land to lose. Stateless entities - religions, ethnicities, corporations - have sort of a secret superpower when dealing with nuclear states. We kind of see that in the post-ww2 era, and the "Cold" War, for the nuclear powers, was a war of stateless factions (a little bit like the European Wars of Religion). A follow on from this is that an unstable land border of a nuclear power is more or less the exact definition of trouble: Taiwan, Ukraine, Gaza/West Bank[a], and going back, the Soviet breakup of course. The real horror story would be a nuclear state run by a stateless minority - like a religious government that hates the bulk of its own people - because they're at least partially incentivized to take nuclear casualties.

[1.a] Yeah yeah I know "Israel Doesn't Have Nukes", but c'mon.

Part of the core problem of our whole economic setup is the widening gap between "where money comes from" and "what do things do?". The VC hype machine makes this gap so large that it can disrupt attempts to design a specific functional requirement. It's way, way, waaaayyy more important (in the near term, anyway) to get the attention of Mr. Moneybags versus making a self-assembling AI wiring diagram. Especially when these systems are so resource hungry - that's a lot of money up front.

We see a flavor of this in the defense industry, where the users of the Product have a . . a whole ecosystem of fusspots[0] . . between them and the money. Say, for example, you get a gigantic requirements document. It might say "You will make the Product System but you must do the work according to Process X, Process Y, Process Z, and any related Process [X.1.b through X.6.z, Z.1 through Z.921, etc]". Then, when you show you have these Processes/Certificates, you get a pile of money. Thing is, it could be decades - or never - before the Product ever sees anything like a user. Every time someone refreshes the Giant Requirements Document, more money gets paid. The actual transaction taking place is Paperwork for Money - so therefore, the red tape is what's valued. Sometimes - almost by accident - a product comes out from this, but very often it doesn't[1], and the whole machine just keeps cranking on.

[0] I am being really nice here, but yeah, without trying to be nice . . there's a TON of corruption, and despite what the NAFO fanboys say, a LOT of procurement officers have their corner offices already picked out at LockBoNorthRay. It's an absolutely widespread practice - every company I've ever been with has BRAGGED about hiring on Procurement Officer X or Y from the latest bagged contract.

[1] Sometimes a Product comes out completely FUBARed[a] and then the techs and engineers have to figure out a way to make it fight. It would all be a hilarious TV show, but in real life all I can see are the dead airmen and sailors that'll be stuck inside when push comes to shove. Well, "shoving" someone other than five Arab teenagers playing with fireworks in an RV, you can fight those guys with a Garfield body pillow and some body odor..

[1.a] But within that ever-shifting requirements spec!

Our local home depot had its entire setup stripped out and redone with self checkout, but the system - in spite of being very glossy and futuristic looking - was designed by someone who had never worked retail before in their entire lives[1]. Moebius loop menus, dead ends that require store login to back out of, infinite loops when undertaking various payment methods[2], and naked windows errors when what looks like Electron rolls over and dies. I haven't seen any one customer successfully use the new setup to navigate a transaction. And it was expensive as hell, according to a disgruntled manager I chatted up.

It's wild. Almost defense-industry-like.

[1] Or possibly ever programmed a computer. I am very possibly the World's Worst Programmer[a], and my completely shambolic first attempt at an Electron-based data entry system with a smartcard auth thingy worked way better than this thing. I at least hid Windows errors for chrissake.

[1.a] I've got that on my business card, along with the incredibly niche things I might actually be able to help you with. Incredibly, I've gotten programming gigs with this.

[2] As shoppers fearfully wonder how many of these transactions are actually posting to accounts.

I'm sorry to say such an adolescent thing here, but, "Duh?". For a delicate regime with many bordering enemies, and with an international nuclear policy in total shambles, it's the most sensible thing to do. By a country mile. If I was Grand Iranian Overlord, you can bet that every national effort would be, right now, to get a nuke as soon as humanly possible.

The Shia/Persian state is a sad little mound in the middle of a Sunni and - before Gaza, anyway - Sunni/Israeli block. I imagine the primary reason for the October attacks was to drive a wedge into that developing relationship, knowing Likud would react very Old-Testament-y.

People will make scary color coded maps with all the Iranian proxies, but proxies are just proxies, and some of those proxies - like in Syria - are as dangerous to Iranian policy as they might be to Iran's enemies. Ask America how relying on dangerous third parties went for it in South America - they might have established a firm anticommunist platform, but at the risk of destabilizing the whole continent.

Respectfully speaking, that sounds an awful lot like someone's last words in the accident investigation, right before "always" suddenly isn't.

The truth is, in a lot of ways, those bombs are Pakistan's bombs, but they're also China's bombs, balancing out the Israel-India axis.

In terms of weapon effect, nukes really are in a whole other category, particularly against invasions, which require concentration of force by their nature. Even a dozen nukes with 1950s delivery systems can make the price of invasion untenable - particularly for democratic governments. And once a country is making them, they're very easy to stash, so you can never be sure you got them all. We faced a similar problem with Cuba, when we learned that they'd stashed possibly quite a lot of FROG launchers in addition to the big missiles.

Chemical weapons are a bit like Age of Sail line of battle, with an awful lot depending on the weather. Biologicals are even worse, since neighboring populations - surprise! - turn out to be pretty closely related, making it difficult to target foreign ethnicities. Humans in general are pretty tightly clustered, although I would imagine one of the great weapon uses for AI systems in this century (or decade) will be biologicals very tightly targeted at micropopulations, very small cell factors (so, with your new AI-designed bioagent, instead of 80% enemy 40% friendly [which was the estimate for an Anti-Black People agent in South Africa] it goes to something like 70% enemy 10% friendly). Both biologicals and chemical can have unpredictable weapon effect, which is the primary reason we don't see them more often. A really great rifle that shoots backwards even .1% of the time would still be the world's worst rifle.

Nuclear states have a form of "super-sovereignty" that effectively makes them a different kind of state. No one's chucking rockets or missiles at you anymore, no one's going to risk an airstrike or putting you on the Regime Change list. The long term effect of holding a nuclear arsenal has some nasty follow-on effects internally, for a nation, beyond just the sanctions and such, but it's such a cost-effective method of deterrence that such factors fade into the background.

I know I've tooted its horn before, but Orange3 is a pretty neat Python-based GUI platform that makes this and a metric buttload of other statistical/ML techniques available to non-programmer types.

Just watch out for null character `x00` in the corpus. That always seems to kill it stone dead.

https://orangedatamining.com/

https://orange3.readthedocs.io/projects/orange-visual-progra...

Damn frickin' straight man. Remember that scene in The Founder? "You're not going to run a hamburger business. You're going to run a real estate scheme that happens to sell hamburgers". With the Welchians, they've distilled it down to very nearly its apotheosis: the old Mob trick of insuring a restaurant, selling it down to the plumbing for cash, and then burning the joint to the ground.

Buyin' you a drink.

STE used to cost $2900 USD for the print copy. So . . baby steps, I guess?

STE, unlike Simplified English or International English or its many cousins, has what I like to call the "ISO Kiss of Death". Basically, with IKoD, you get a cabal that takes over a spec, and then it defends the spec like a kung fu master defends his secret style. Thing is, the cabal also mandates how documents get written, so now you have to go to the secret Kung Fu School to learn the Secret Style if you want to, yknow, deliver a product.

This can go on for decades, multiple decades, before something maybe breaks the chain - some busybody in procurement, or a major national embarassment, or an overactive journalist revealing how making a PDF file managed to cost two hundred million dollars. Or someone raising their hand and saying, um, sirs, very sorry to interrupt, but this is all common capability in the Lands Outside the Secret Temple.

Fighters from secret temples never amount to much.

I'm still in that niche, and I hope I'm not being Debbie Downer, but no, the software is still horrible.

The only movement that's happening today is where individual ERP/PDM/ILS solutions are beginning to just incorporate the tech writing tools wholly into their ecosystem, rolling them up into SAP/Epicor/TeamCenter/Aras/etc. Or even into the CAD system[1]. Doing S1000D in Program X? Whelp, better get TeamCenter spooled up so you can do the writing and publishing.

What does that do for interoperability? Nothing good! But it doesn't matter - S1000D instances weren't interoperable to begin with. The one good thing from this, is that it might - and I do want to underline might - give the pubs team a somewhat better chance of optimizing their data flows from other business systems. Yeah. Might.

All this crap makes me think of Daniel Dennett's famous essay on Chmess. How "chmess" sucked in a whole generation of chess scholarship, with no one stopping the train for one hot second to ponder that they're blowing their entire careers optimizing something that never happens.

[1] Oh yeah, CAD for maintenance manuals. The engineering design system, that's a great place for parts lists.

I move in these military/defense spec circles rather a lot, and yeah, ASD has come light years when it comes to ASD-100 (STE) availability. Cruise this thread and you'll see quite a few of the horror stories users have had just getting their hands on a printed copy of the specification. Why, you might ask, why in the dickens was this arrangement so resilient? Whelp, it goes back to the governance of these specs, and the ugly truth is that if the national military sits out the technical steering, and then the big OEMs sit out the technical steering, the only ones left are the existing software vendors. So that probably tells you all you need to know about how that goes.

I'd also like to give a shout out to TechScribe.uk, who sells a LanguageTool package to use STE as a LanguageTool grammar and language set. RedPen.CC also has some STE rules you can plug into the linter. RedPen and LanguageTool, you can integrate those with Visual Studio Code or another standard text editor. Of course, if you're tasked with STE, then you probably also have the budget to go to the Big Boys, Etteplan (formerly Tedopres) and those fellas, who are in the spec governance up to their elbows. There's a reason licenses for aerospace techpubs can easily clear 50k per seat per year - and the reason is crap like that.

The pay scales are locked down by two primary components: 1) the project contract document locks charges pretty severely for the project, to the point where a single high pay scale putting in hours will probably require a revision to the charges, but more importantly, 2) the owning Big Defense Conglomerate has armor plated pay ceilings for non-managerial employees - non-financial technicians are interchangeable cogs, that's the doctrine, regardless of any particular realities of skill shortage or availability.

Disregard what the Overlords might tell you; 2 is way more important, because when it comes to 1, the military program offices are VERY open to revisiting the charge schedule, if it means they get transparency and a better chance of something that might work. Those non-managerial pay ceilings were laughable pre-Covid, and now it's gotten just surreal, still based on a flat national average of what they consider the job role to be, something that's also gotten surreal, with Kinkos employees being included in sw engineer pay codes.

It's worsened by the fact they can't really take advantage of remote workers, somewhere with less CoL, in Alabama or Pakistan or whatever. Everything's on site, and it really does seem like every defense company I've ever worked is sited in some random shitty place where the local economy - if there is one - is composed of the company and three Applebees-like chains that service it.

In the defense and aerospace industry I've known - and continue to know - a number of engineers and specialists who do this. Leadership will often turn a blind eye, because the particular skill is so specialized, they'd need to give up a product line (or the whole business) if they terminate the guy. But, this being a very sick industry, they can't actually pay the guy more, so you get these terrible arrangements.

This is interesting. LibreTranslate has been my go-to offline translation system for a few years, but I need an upgrade. This looks like a decent candidate, assuming I can make it work with what we got.

Google Edgewood Arsenal and the BZ munition.

Long story short - just like with gay bombs and all these other sorts of Cold War esoterica - is that humans are really complicated, and war is even more complicated. Turns out making people gay or psychotic - even if the chemicals did work at doing this, which is dubious given their tenuous understanding of what "gay" or "psychosis" was - doesn't actually reduce their warfighting capability. Sometimes, it even made a tenable tactical situation much worse.

Now, when it came to making people think they were going crazy, that's where these substances really started to find some traction, but that wasn't a military application. More the territory of the CIA headshrinkers.