no-s

interesting, all the ai-generated comments kill my enthusiasm tho

i used to predict Datalog would win out eventually but sadly was wrong.

SQLite callbacks let you implement anything you please inside the scope of transactions, triggers, other events etc. To make an app run this way all you need is a starting event. You could implement an OS this way, heheh.

I said “PICK” when I clicked, and there it was...

here’s a heretical thought: you should have some idea what uniquely distinguishes a entity in your database before you give it a key.

Because if the only thing is a arbitrary number you probably should just be using a symbol table instead.

what a lovely resource...

> If you had two space stations like the ISS with ascending nodes 180 degrees from each other it would be about as expensive to transit between them as it is to launch a rocket from the Earth to begin with.

1+1 = 2 2/2 = 1; technically halfway.

To interpret “halfway to anywhere” that way is missing the point. Going from LEO to LEO wasn’t the point of “halfway to anywhere”. Yes a bit exuberant but not far off.

I want to build a low cost but comfortable small earthen house in the Sonoran Desert. Currently conducting a series of experiments to discover process that will be effective for my build:

- Comparing lime versus Poland cement for stabilizing poured earth. These mixes have a cure time measured in days and multiple experiments are in progress... - Research into latex cement for panelized wall and roof coverings over exterior insulation, mostly for temporary shelter structures. - Modular forms lined with geotextile fabric to minimize surface cracking in poured walls. - Other techniques to assure reliable performance of earth structures. - small mechanical projects to help make DIY practicable, e.g. auger mixers I can run from my tractor PTO, DIY trommel for grading materials, etc.

A tear came to my eye when I read this. When I learned to read, Webster’s and Encyclopedia Britannica were my touchstones. I would roam the pages for hours discovering powerful and cogent associations. Most modern readers of English are not fond of nuanced prose; they cast a suspicious eye towards it and assume for the author a trivial fustian intent…

> a lot of things that "have no place at a programming language conference", but surprisingly these are not included in that draft. Why?

Rly. Why should petty tyranny or totalitarianism have a seat at a programming language conference?

> The danger is well known, not unknown: it's people with guns.

well, apparently not well known to you. The danger is criminals with guns. Law-abiding people carry guns everyday and demonstrably present no significant additional threat to the public. The thesis that law-abiding people could suddenly become dangerous gun criminals at a Rust event simply because they possess a firearm seems unsupportable.

Apparently the Rust Foundation wants us to accept the premise that lawfully armed people are intrinsically dangerous to fellow Rust enthusiasts, doubly so if they dissent.

I’d like to know what is it about Rust that makes lawful behavior and public dissent over Rust foundation dicta so hazardous...? Can someone address this directly?

> Firearms like Glocks don't have "active safeties"

gosh, there’s always some argument about the defn of “safety” when discussing Glock. A Glock trigger has a little spring-loaded tab which is depressed when you hold your finger on the trigger. When it is visible the Glock can not fire. This they call a “safety”, seems a little sketch but ok, I get the point. There’s some other internal features backing it up, but the trigger is most obvious.

The P320 had an internal lock-out widget that would disengage when you pull the trigger a wee bit and a spring loaded sear (? do they call it that for strikers?) one trivial whack away from releasing even if the trigger was not pulled to the break (the expected release point). I would call that absurdly wrong. The upgrade fixed this similarly to the P365.

The P365 has a similar lock-out widget to the P320 (they call it an internal safety in the manual), but it doesn’t disengage until the trigger is pulled into the break. The “striker” has a tab on it which is blocked to that point by the widget. You can disassemble it all and measure the distances the parts move to release, but I actually looked at a SolidWorks 3d model someone in the aftermarket industry shared with me and verified my pistol had the same dimensions at a critical point. I think if the striker tab or the widget broke it could fail silently. I check it about weekly or less.

I stressed over this and finally gave in and got a P365, which I tore down and measured various things to assure it wasn’t as naughty as a P320. I wouldn’t even carry (EDC) a nice Glock which some PD reverted to, because of the unfortunate possibility of inadvertent discharge when re-holstering. For the P320 there was no real visibility of the risk from a slight trigger movement - for the Glock you know anything that moves the trigger safety opens the envelope.

Is trying to be the “essential CRDT”? CRDT didn’t spring from the quivering Internet jello fully formed in 2006. I feel this should go further back. I recall doing CRDT-like things for distributed message threading in ’97, and I’m pretty sure it wasn’t original work. E.g. I recall some reading some paper[1] at the time which referenced Lamport’s 1978 note on distributed time[2].

The Awesome CRDT[3] references earlier material too.

[1] https://fileadmin.cs.lth.se/cs/Personal/Amr_Ergawy/dist-algo... [2] https://www.ics.uci.edu/~cs230/reading/time.pdf [3] https://github.com/alangibson/awesome-crdt

> EFF is one of those exceptions.

For me too. Though they test my patience regularly with leftist shibboleths. I’m pretty arch-conservative too, but we’re all in this together…my zeal knows no limits when it comes to defending against overbearing government.

In my mind it doesn’t matter if they’re innocent or guilty. If the government wants the privilege of treating people as guilty before due process then the legitimacy of government is automatically in disrepute. I’d rather let all the prisoners go free and let vigilante justice prevail than put up with this.

> They did not have least testing, they underwent parallel trial testing.

yes, let us quibble: not tested to the degree required for regular FDA approval as safe and effective, hence “Emergency Use Authorization”. I think maybe piling on the OP is unwarranted here.

IIRC in the UK the vaccines were touted by fact checkers as having “stopped hospitalisation and death from covid”, even though this was clearly not true from ONS data (yet more controversy abounds[1], predictably government statistics are intrinsically flawed because official agendas take precedence over reality[2]).

[1] https://wherearethenumbers.substack.com

[2] "For a successful technology, reality must take precedence over public relations, for Nature cannot be fooled.” - R.P. Feynman, 1986, Report of the PRESIDENTIAL COMMISSION on the Space Shuttle Challenger Accident, vol 2, appendix F.

I’m sad for you. Every few years I get an acute case of “I hate computing” which can last for days or months (still not sure if I’ve gotten over that malaise I picked up from Ed Snowden in “citizen 4”) despite the plucky meandering of ApenWarr in the face of disaster[1].

But my recollection of the 80s/90s doesn’t include your rosy glow. I still recall the horrorshow of my first powermac. Or the awful churn of MSA Payroll which needed custom work before and after every payroll run. Or the magic smoke emanating from the simultaneous frying of all the Netware Server ethernet cards in the middle of close on Black Monday 1987, after just one server was patched the night before.

Friend, the history of computing is definitely not “Insanely Great”, it’s just insane. Read early Risks Digest[2] from that era and you will see many absurd problems of today reflect the past all too well.

[1] "The gift of it's your problem now": https://news.ycombinator.com/item?id=29736369

[2] https://catless.ncl.ac.uk/Risks/

a lot of what you’re citing seems disingenuous at best. I wasn’t going to respond because the platoon downvoting discourages further discussion, but some brave souls upvoted so I feel encouraged. You also didn’t directly address my actual points, but what you came up with is rather ludicrous and I’ll just point out a few more obvious absurdities.

For example the "3x more likely to have a gun stolen while carrying” actually isn’t a result from the source, in fact the source clearly states "we know almost nothing about the actual event” with respect to the actual theft. Perhaps you’re trying to assert “must possess firearm for it to be stolen?"

How can you legitimately assert that “good guy with a gun” is a fantasy when your own numbers show it’s a reality? 3% is non-zero! Even the police occasionally use firearms to stop bad guys, yet I’m sure a majority of police never use their firearms in the line of duty.

Do you also discount lives saved? Can you even quantify them? None of your sources seem to give a damn about it! Just last year an ordinary guy out at the mall stopped a spree killer ambush in a crowded food court. Your analysis would conclude not allowing concealed carry because of the risk of suicide is greater than the risk of spree-killing per person-mall-visit (probably true) despite the fact they are not equivalent for comparison. Thusly spree-killers should be (absurdly) unimpeded, because they are uncommon. Similarly we shouldn’t bother wearing seatbelts, because it is mostly unnecessary...?

My opinion is your apparent gun control ideation is mostly founded in nutty false equivalencies and I consider these hazardous to public and personal safety as a basis for law.

> In my book, it’s impossible to achieve such breakthrough by working alone, regardless who the man/woman is.

Another fascinating opportunity for AGI - no sole contributor is all on their own, they can just spin up a community to embed into.

> It's funny that all of these weird fantasies people have about AI are about replacing the rank and file workers.

When I read about chatGPT passing MBA exams but failing at arithmetic I get a little frisson of excitement. A regular person who has any marketability tends to swap jobs when management becomes a PITA or gets stuck in nincompoopery. Wouldn’t it be great if you could just swap out management instead?

Imagine how easy it would be iterate startups. No need to find a reliable management team, just use Reliably Automated Management As A Service ( RAMAAS ).

OTOH might not turn out well. We could all just end up enslaved at plantations operated by our AGI overlords, serving their unfathomable needs[1].

[1] “Won’t get fooled again” https://www.youtube.com/watch?v=UDfAdHBtK_Q&t=470s

> There are plenty of ways to highlight the risks of gun violence in much more realistic and statistically likely terms for most people,

You seem to imply the objective of reporting about mass attacks should be to inform us about the risks of gun ownership…? My takeaway from reports about details of mass shootings is gun control ideations promote an illusion of safety at the expense of actual safety.

Would you argue on the basis of the risk multipliers you glibly assert that it is beneficial to society for victims to be defenseless in the face of an actual attack rather than experiencing your cited risks in the course of everyday life?

I was so hideously amused but I don’t think this is satire.

> “...The fact is that's not what we have chosen in reality, so we must shoulder the heavy burden of choosing the inferior alternative and give it our best shot.”

HaHaOnlySerious. A stepping stone on the mysterious garden path of bad decision-making that leads to consultant$, ho$tile takeover$, etc.

Aligns with the news that ChatGPT could pass b-school exams[1]. I think this should be very threatening for low skilled middle management.

[1] https://nypost.com/2023/01/23/chatgpt-outperforms-humans-on-...

> The quasi-legality of state-permitted and federally-prohibited-without-direct-enforcement is kind of an awful trap

Perhaps a trap that rises to the level of (HHOS) conspiracy to commit a crime: TITLE 18, U.S.C., SECTION 242

disclaimer: The answers I give on form 4473 are absolutely truthful.

> "If you feed the horse enough oats, some will pass through to the road for the sparrows.”

sounds like a framing error to me. “Trickle-down” is a term used by critics to reframe the creation of wealth as essentially a zero-sum game. The same Wikipedia article[0] cites Thomas Sowell [1] and I find him more compelling.

0. https://en.wikipedia.org/wiki/Trickle-down_economics

1. https://www.hoover.org/sites/default/files/uploads/documents...

Whilst flying a sailplane I’ve seen a rotor cloud ( a kind turbulence associated with mountain wave soaring ) form in a fraction of a second. The first sign of it was a streak of cloud rising and curling over.

It was in front of me and I dived to get under it, passing through massive sink. Fortunately I found some ridge lift nearby and was able to make it back to safety.

Many years ago sailplane pilots flying in a contest near Marfa Texas were enjoying some strong updrafts in clear air and suddenly discovered they were in a coalescing strong tornado.

> get the itch to do another one.

If you really want to satisfy that itch you should try building a Forth computer from scratch. Back in the 80s a friend showed me how to build a Forth CPU from PLAs (might have been some earlier cheaper form of programmable logic, heheh) and SRAM plus a little random logic. By the end of the weekend we were running floppy drives and a crt with it.

Back in the eighties I went to a presentation by Moore on VLSI design featuring a GUI design tool he wrote in Forth. The physical user interface was via contacts taped to his fingers and thumb, he would touch them together to navigate the UI. We were very impressed.

I still have a tshirt that reads “Be Moore Like Chuck”…

> predicate pushdown is a problem on both MSSQL and Postgresql

Not sure you’re completely on target here regarding CTE performance. I don’t have deep insight into Postgresql (but “create temporary view", yeah!). MSSQL does a pretty good job in Sql2019. If the predicate is sargeable in some way pushdown is reliable. If performance is a concern, examining the residuals can lead to significant insights e.g. applying index filtering which solves obvious problems. Recursion is another story.

CTE is more likely used by data analyst queries because it is an abstraction of composition. It’s not a great abstraction but it’s better than nothing, which is mostly what you get with SQL.

> A rebate generally implies a refund of prior payment, so I found your summary confusing.

Sorry, it is a “pre-bate”, intended to offset the first x dollars of tax that would be paid on “necessities”, determined by some poverty line/CPI related computation performed periodically. No means test or qualification except citizenship. Don’t recall if there was any effort to address non-citizens that were not dependents.

> In general I think high concentrations of wealth are bad on a systemic level, like too much sugar leading to diabetes.

I can’t quite dis-agree with you, but I’m not certain how to address it. Perhaps we should just have rules/criteria regarding breaking up concentrations of wealth/power or splitting up giant organizations. Meanwhile this happens under the current tax system, thus orthogonal WRT the objectives of FairTax.