You missed the step where TPTB—after delivering on their master plan to destroy Uber from the inside out—send out Luigi to collect their rightful graft from all the corrupt^H^H^Hgreatful taxi services, taxi unions, and local politicians and governments to recover their original investment of $27M, their opportunity cost of lighting a match to their own Uber stake (having along the way become potentially worth quite a bit on its own terms), all their time and careful, world-class plotting to destroy the apple from the inside (without making any of this so obvious as to get their asses actually sued for breach of fiduciary trust), Luigi's salary, and Luigi's nunchuck stipend (got to keep up with the Furios).
Tear a page from the Hermit Kingdom, and what you end up building will have the same level of intrinsic merit: a privacy shroud that could be broken by an ambitious elementary school kid.
The guest, Li-Huei Tsai, director of some august learning and memory institute at MIT, eventually confesses that transfer from mice models to humans in this line or research has about a 1% success rate.
it will only be used by the opposition as a proof that you are always wrong
Proof used to be universal. Now speaking to your base involves one form of "proof", while conversing with educated humans involves another.
Many people don't seem to mind being called base, because it implies staggering levels of power (alongside staggering levels of stupidity). Unfortunately, Donald seems to have lost the tired-of-winning remote engine starter fob under some Whitehouse sofa cushion, since replaced, and now all that staggering power remains parked in the driveway.
Looking stupid.
And not making enough noise to make stupid look awesomely impressive.
Though it sure looks like it could make that much noise, if ever Donald found the keys again.
Wow, IBM attempted to throw more hardware at the problem, and actually succeeded.
Downside: requires more hardware.
It probably does decrease average training latency at scale (not counting the shovel time invested in sunk cost), but I'd still hesitate to call this an advance.
The Greeks and other ancients had wonderful methods of memory that were very impressive and lamented to the death of these when written language began to flourish.
Those 'impressive' Greeks would fall off their pedestals if they had half an inkling of the amount of knowledge the average broadly read and well-informed IT geek of today carries around as a matter of course.
The Library of Alexandria had somewhere between 40,000 and 400,000 scrolls.
As of 8 August 2017, there are 5,456,325 articles in the English Wikipedia.
I suspect that 10% of these articles would be as large as any dozen ancient scrolls.
I figure the keyboard and the piano will walk into the sunset arm in arm, i.e. not any time soon.
The recency bias is pretty simple. Because it's easier, we're inclined to use our recent experience as the baseline for what will happen in the future. In many situations, this bias works just fine, but when it comes to investing and money it can cause problems.
Well, I suppose there are worse problems in life than paying 20% more for 20% less because of an edge case.
Unless you make it a habit, and it begins to consume larger fish, like your 401(k).
I glanced through Sarah Mei's Twitter page, and she's full of shit. She seems not to get why women in tech might not be evenly distributed.
* Suppose you have a culture that hires based on personal referral. (It's usually one of the best ways to go.) * Suppose your culture starts out with a male nucleus. * Suppose your male nucleus mostly has male contacts.
You're gonna get a mainly male culture.
Companies don't hire the best candidate available. Companies hire the candidate for whom they have the most confidence of strong performance, meaning that the route into the door matters a lot. Applicants at large will not be given equal shrift to applicants with a strong, internal referral.
From that starting point, the organization is subject to network effects, none of which need to be intrinsically biased in order to lead to a biased outcome (as determined by simple headcount).
One can argue that the sorry state of women in technology justifies taking active measures against the default behaviour of your (potentially) gender-neutral starting point. One can't argue that failing to take active measures automatically incriminates your starting point as gender discriminatory.
In Sarah's world where water isn't wet, and laudable corporations seek the best candidate while paying no attention to existing network effects, you can draw these conclusions, loudly and with no nuance, should it serve your purpose.
I'm not saying that innocent bias doesn't coexist with toxic bias. I am saying that presumptive guilt is an extremely dangerous tool as wielded by a small, angry imagination.
One of the most maligned aspects of the TRS-80 is its cassette loading procedure. Interestingly, it is a lengthy and skillfully designed piece of coding, a victim of a combination of poor hardware (an inexpensive cassette recorder), the inclination personal computer owners have to purchase the least expensive tapes they can find, and the lack of foresight on the part of the engineers designing the routines. But there's no question that with a good tape recorder and reasonable tape, it works well.
...
Overall, these routines give the appearance of being reasonable and reliable, and they should be. What, then, gives rise to the tape problems? Mostly the timing loop in the 0235/0241 subroutine. The values placed in the B register at 0248 and 024F are too short for low-grade audio processing. Simply stated, the audio waveform coming in from tape 'rises' too slowly for the fast bit-check loop at 0251 to catch. A 'one' might come through, but it comes through too laggardly for port FF to have flipped into place.
I found that quickly because I can still recall the entry point 0x1BB3.
I got that wrong in my mind, didn't I? It takes you back to the call point. Oops. Difficult business, this is, dredging up memories from once-upon-a-forever-ago.
I'm pretty sure I wrote that program for maximal obfuscation, so there could have been another for loop in the subroutine, and I really was ping-ponging wildly.
My high school purchased the original 4 kB version, and then upgraded it to 16 kB the following year. The school generously allowed me to take it home on weekends. There were two Apple IIs in the lab as well, with actual floppy disks. I chose the TRS-80 simply because I could get more time. The Apple IIs were busy playing games most of the time. Did anyone learn to actually code on those machines? Not that I noticed.
This was my first actual computer. In my first week of programming, I tried to write:
if 0 <= i < 100...
I was shocked when this didn't work as expected. Man, I thought to myself, these things are even more brutally literal-minded that I had ever imagined, and I had already been reading about computers for years.
After that, I constantly had at the back of my mind: just how broken is this language?
Well, you can write a for loop. Inside the for loop, you can call a subroutine. Inside the subroutine, you can write next. This takes you back to the for loop. Then inside the for loop, you can write return. This takes you back to the subroutine. Only don't do return again, your call stack is empty. You need to get out again with another next statement. Yes, I actually wrote this program using the TRS-80 BASIC.
It wasn't long before I wrote a skeletal Z80 disassembler in BASIC and then reverse-engineered the ability to load machine language programs directly from cassette tape.
After that, I didn't program in BASIC very much.
I still have a TRS-80 reference manual with the full schematics.
No, my idea doesn't work, because the Mafia can do the same thing in reverse: gather up all the receipts associated with "paid" votes, then randomly test ten (a $10,000 cost-of-doing-business fee), on penalty of worse-than-death.
I think that would reduce the enforcement cost enough to turn paying for votes into a cash-flow-positive business model.
Bear in mind that delivering on penalty of worse-than-death is not cheap (either in time now, or potential for time later). If all the rabbits are trembling enough, you won't need to do this.
With suitable encrypted signatures, you could set up the receipt so that you had to pay a $1000 fee to check that your vote was accurately counted, and if it wasn't, you are awarded $1,000,000 in compensation (to deliver the sting where it matters, each election oversight group could be organized like the unlimited Lloyd's "names" of old).
Why would you do this?
Well, there's a problem with ubiquitous, easily verified receipts that's usually covered in the second lecture of Public Administration 101.
I don't presently feel like Googling that for you.
The entire brilliant new thing about deep learning is that you can build an entire machine translation system from fucking randomized matrices, all the way up, where no-one got to decide anything.
Hand-crafted rule-based systems present thousands of opportunities for power-mad silverbacks to dicker to their own advantage (see Swamp, The).
But with deep learning, you bootstrap the system with massive artefacts extracted from the real world (the training corpus) and even if you wanted to dicker with the artefact, we've got barely the first clue about how to tilt the artefact—bear in mind that it's very, very big, with a low center of gravity—so that the machine learning algorithms respond in a desired, predictable, stable way (that isn't entirely upended by the next trivial dicker).
Wake up and smell the bacon. Gradient is out there, and mankind no longer sits at the top of a micro-manageable food chain.
This has always been true in small corners of human affairs. At the end of the day, it really doesn't matter who invented the calculus. For the most part, the calculus is a Platonic good.
Networks can lie across a fairly wide swath of the Platonic–political spectrum.
We kind of lucked out with the Internet. ARPANET desired a certain form of resistant against politics (i.e. the backbone coup) that aligned with the individual's preference against being controlled (they didn't foresee, I don't think, how soon social media would become a larger stakeholder than the Pentagon, or more of the ribbon-breasted control freaks would have popped out of the ARPA-oversight woodwork).
If our explicit goal is to tilt way over to the Platonic side, it's not like we have a huge number of dials to bicker about, anyway.
Resource management requires some kind of accounting system which identifies endpoints (bandwidth is neither infinite nor free when push comes to shove). I don't know whether our anonymous micro-currencies are up to the job yet, at such enormous scale.
How does one respond to a DOS attack on a fully onion-routed fabric? Sounds like a tough problem. If it's not onion-routed, there's clearly a small privacy leak that could be exploited by nation-state agencies.
Real problems.
If this ends up becoming a voluntary network (you can join your node if you want to), then like all good libertarian systems, the primary vote is conducted by the pitter-patter of many feet.
In such a world, when the technical committee gathers together, they are going to look around the table to see whether the assembled group has the competence and credibility to prevail in a vote of the feet—because otherwise they're just squandering their time and reputation to get involved in the first place.
So there's you final answer. At the outside boundary condition, we all decide.
Internal to this, Newton will either decide to work alongside Leibniz (good idea if he wishes to succeed) or not. So, yeah, if your amygdala is so inclined, there will likely be a spot of Alpha Geek Mean Girls during the voyage, that you can happily point to forever after as responsible for any lingering imperfections of Internet 2.0.
That's the ultimate in couch-compatible issue trackers: 999 valuable reforms all blocking on "solve human nature". Congratulations, you are now the proud owner of a labyrinthine saddle point that stretches as far as the eye can see in 200 fucking dimensions.
If it weren't for the giant "who decides" monolith erected at consensus centroid of Saddleplane Peak Perplexity, no two people wandering alone in that vast undulating outback would ever meet up to exchange bile.
Well, sure nice to see a human face every so often. Best of luck to you. Me, I'm heading thataway...
It screams not for business or adult use or Christian use.
Let's also mention your invocation of "all Christians are created equal".
Publicly they move as a group (sort of, sometimes), privately they bicker lick hell. But they sure love to sell "we are all one" when it allows them to assert their particular brand of moral spray paint.
Catholics and birth control are high on the list of All World hypocrisy gradients.
Let's consider another case. On the plus side, they got Neil Gorsuch. On the down side, they just sold their collective souls to a devil who is a real life cartoon figure possessing real life nuclear weapons, but bereft of any emotional, spiritual, or moral hesitation whatsoever to turn the other ass cheek at the least provocation.
(What would Jesus do? He would run. Run, run, run far away. Nice guy, backbone not included. Think about it. If he is still in the business of carrying people along the sandy beach of life, he lately seems to be pointed in the wrong direction. Hey, Jesus, if you're listening, get yourself a moral compass and turn this damn ship around.)
Meanwhile, the Victorian era just called. They want their table drapes back.
Going ape shit over the FreeBSD logo—to give you troll post any credence at all— is a form of triviality porn, a favoured "gotta do it my way" micro-penis performance art of legacy privilege.
Relative importance of the FreeBSD logo semiotics compared the daily news cycle: about 10^-12.
The mascot raises all kinds of questions for people. Rather than having people focus on the benefits of the OS the first thing they see is the mascot and thus the OS is dead to them before they go any further.
I would just love to see you in action in a focus group round table discussion.
Why are all these other people here? Trust me, I've got it covered.
Most of the comments I skimmed are missing the point.
The real problem is that even scientists with the best training and the best intentions wind up committing a certain amount of p-hacking subconsciously. Just a simple data exploration to decide post hoc whether any collected data is corrupted or implausible, and you've already slithered one toe across the p-hacking line.
When p-values gate publication, and publication gates promotion, you create a severe moral hazard where many of the scientists you end up promoting lie on the bottom half of the curve in self-policing their accidental p-hacking. The guy with the penchant to do slightly more irregular experiments, which require slightly more data cleanup, seems to get slightly more published results. Ba da boom.
p=0.005 would put a pretty big crimp in this effect.
Of course it doesn't solve the larger problem. But good golly, first things first.
We also know from replication efforts that p=0.05 is allowing far too much crap to float over the gate. p=0.005 probably gets us closer the crap level we naively assumed we'd get when we originally rallied around p=0.05.
Probably the increased use of computers hasn't helped matters: even accidental p-hacking with pencil and paper is hard work.
The purpose of my post was to troll-shame the one-trick ponies, which was so easily accomplished, I managed to press "submit" without even stopping to think.
If teams with women have a different pattern of returning hat-in-hand for a subsequent round (perhaps with better foresight and anticipation, hence less desperation and better leverage) then the VC could very much end up munched in the ass by imprudent gender equality.
The correct metric here is return to VC investor. Perhaps teams with women are less easily shaved in a subsequent down round. Then it would be rationale for VC investors to eschew EQ on the business-end of the puppet strings.
Missed a trick, being too accustomed to almost making myself comprehensible.
What follows is a post-historical Chicago Manual compound-modifier satirical-punctuation meteor shower.
I'm so ashamed. It would have served as the perfect foil for my subsequent riff about withholding the verb for too long.
Actually, the problem here is that the context is cold. It would be perfect in this more extreme form after warming the reader up with some sentences thoroughly stilted in the other direction. But in this piece, this sentence is positioned in the transitional foyer, and the extreme form is too abrupt.
On second glance, my shame decays into a shrug.
With this kind of humour (supposing there is any) it's often useful to ask yourself along the way, "well, who would be the perfect reader for this piece?"—because if you can't name even one perfect reader, it's likely that you have phased entirely out of the plain of comprehensibility.
Scala-loving Dinesh Chugtai would be a good choice. He'd definitely have laughed at my turn of phrase "Pandit Chin Thumb". And he'd also at least detect my scatological / linguistic / functional programming triple-entendre at the very end.
What follows is a meteor shower of post-historical Chicago Manual compound-modifier satirical punctuation.
Configure ears in the upright, locked position, and proceed on impulse power only. ___
A section of workspace in the circular, Norman Foster -- designed building is finally move-in-ready
I'm pretty sure a fullname–verb ndash is properly ASCII-rendered as Norman_Foster-designed.
This could have been one of the best double-pips ever, if our submitter were irony-enabled about Slashdot's Flintstones–Jetson retro-chic electronic heritage.
Then the headline could have been: How Jony's Norman_Foster-inspired monument to Pandit Chin Thumb ransacked a Death Star fashion parade.
(Unfortunately, my sense of humour often makes the highly mistakable[*] blink--and--you--miss-it-twice "whoosh, whoosh" sound of a side-holster unfriendly dipole photon saber.)
[*] Do notice my clever "-ly" guru override.
Rewriting the full sentence to also address the weird colon:
A section of workspace in the circular, Norman Foster–designed building—sliding-glass doors on the soundproof offices, a giant European white oak collaboration table, adjustable-height desks, and floors with aluminum-covered hinged panels, hiding cables and wires, and brushed-steel grating for air diffusion—is finally move-in-ready.
This sentence, as properly recast with the long, orthodox, purely phrasal appositive, also has its own, internal humour, with an olllllllde-English Teutonic tonsil-vibe inherent in the long-withheld sentence-final verb.
Erlich: Richard, if you're not an asshole, it creates this kind of asshole vacuum, and that void is filled by other assholes, like Jared. I mean, you almost gave him shares. You need to completely change who you are, Richard. A complete teutonic shift has to happen.
Richard: Tectonic.
Erlich: What?
Richard: A "tectonic" shift is the earth's crust moving around. "Teutonic", which is what you just said, is an ancient Germanic tribe that fought the Romans. They were originally from Scandinavia...
Erlich: Stop it! Stop it. You're being a complete tool. Right now, I need you to be a complete asshole.
Come to think of it, too much withholding the verb was the subject of a bitter bear-cam quarrel in The Wolf of Wall Street.
Perhaps the colon is well suited to this innovative folding-function, after all.
I'm not trying to be difficult or argumentative, but MS Paint always seemed like such a crappy program.
I'm not trying to be difficult or argumentative, but Grandma always seemed like such a useless person.
Yeah, you're right, but she doesn't take up much space and she's the only person in the house who can consistently walk into a strange kitchen and crank out perfect fettuccine noodles without having to hoist an unfamiliar stand mixer onto a foreign countertop, and then dig through all the cupboards for a noodle attachment, if the cheap IT bastards even have one lying around.
But she's really old and out of date and doesn't even use ribbons!
Well, suit yourself, but I'd rather keep the old bird around, in the handy drawer in the closet under the staircase, where she's been living quietly without making a fuss for the last 35 years.
According to coworker who is a martial arts expert, losing a pound per week is a sustainable over the long term.
Yes, so long as you aren't simultaneously sustaining any other thing. Like a day job.
I'm joking just a bit, but the word "sustain" is commonly abused in exactly this way.
Weakly sustainable: when just this one thing can be sustained.
Strongly sustainable: a member of the set such that all strongly sustainable things can be sustained at the same time without surpassing the labours of Hercules.
Whenever someone says to me "sustainable" regarding a personal resolution, my first (usually silent) question is: have you ever given one hour notice at work, and then set foot in Tibet the very next day?
Because, if so, that's just a steaming pile of dedication porn.
Then again, suggesting there are "no real options" sounds like a setup for a No True Scotsman fallacy, so I'm not sure that you would have been able to suggest anything to his satisfaction anyway.
His satisfaction is quite irrelevant. Unless you believe that Any True Scotsman would faff around indefinitely to shave one more nickle off the purchase price.
What matters here is his prospective utility: his net upside after the huge investment to research the alternatives, reinvent his established workflow, learn about all the new nits and gremlins, flawlessly administrate his custom stack of validated alternatives, resolving interoperability difficulties with his contacts and clients, etc. etc.
About 10% of all open source zealots really ought to check themselves into Faffaholics Anonymous.
If your wife tells me you drove off to work an hour ago, and your commute is half an hour, it is reasonable to assume you actually are at work.
The thing is, you sign up as a Republican, and on the way to work, the company who employs you suddenly turns into Trump University, and then one day, like Morning Joe, you announce to the public: "you know what, I'm just not into working for Trump University even though I'm going to continue to be a Conservative".
Now, it is true that a blind partisan Democrat can be assumed to still be a Democrat ten years later (even if they did nominate loathsome Hilary), but do we really know the guy ticked off "blind" and "partisan" and "until death do us part" on his original Democrat Vow of Perpetual Allegiance?
I'm sure there are many Obama Democrats who signed up because of Obama, and then checked out of politics during the last election cycle when the Dems nominated a previously-owned, pant-suited, Wall St toady (if they hadn't already checked out halfway through Obama's first term, when he proved to be a colossal disappointment in standing up to the financial sector).
Furthermore, we have strong evidence of the great Democrat check-out in the form of Trump's ghastly victory. We're not even sure he could have won this election running against a ham sandwich, although he did win this election running against "crooked" Hillary. Even though Trump personally thinks Hillary is infinitely worse than a ham sandwich, he still finds time to brag about his victory as a meaningful achievement.
Hillary: high bar or low bar? Pick one.
For myself, the most effective thing I could do politically would be to register with my regional party on the right (I'm not American) so I could help nip against the darling single-issue candidates of the Christian Right in the bud at the nomination stage. This affiliation wouldn't be too hard for me to pull off, because I actually believe in the good half of Libertarianism—compassionate Libertarian wouldn't even be that far from the truth (though my opportunities to fully align myself are thin on the ground).
Then I would be, by American standards, a registered Republican. I'd be happy enough and it would all go swimmingly until Scaramucci shows up, and only then, I would totally blow my cover.
There are hapless Greenpeacers on the left who say equally ridiculous things (splitting desk with head ridiculous), but they're into Volvos and macrame and other mostly harmless things, whereas Scaramucci is a spoiled child of privilege, and I just can't stand fucktards who defend their ridiculous views by pointing at their bank accounts—in a disgusting show-boat of glitzy ad hominem argument (the good half of Libertarianism believes that all men are created equal, the bad half of Libertarianism believes that all dollars are created equal; a die-hard Conservative is someone who conveniently neglects the difference when ideology hits the ballot box).
Rational choice theory therefore had to be elevated from an empirical theory covering certain empirical contexts into a normative theory of the proper operation of the human mind itself. It had to become a universal philosophy. Only then could it justify the US' self-assumed global mission of bringing free elections and free markets to the entire world.
I didn't like this article much, but it did make one or two good points. Rational choice theory has long aligned itself with reductive analysis.
You missed the step where TPTB—after delivering on their master plan to destroy Uber from the inside out—send out Luigi to collect their rightful graft from all the corrupt^H^H^Hgreatful taxi services, taxi unions, and local politicians and governments to recover their original investment of $27M, their opportunity cost of lighting a match to their own Uber stake (having along the way become potentially worth quite a bit on its own terms), all their time and careful, world-class plotting to destroy the apple from the inside (without making any of this so obvious as to get their asses actually sued for breach of fiduciary trust), Luigi's salary, and Luigi's nunchuck stipend (got to keep up with the Furios).
Tear a page from the Hermit Kingdom, and what you end up building will have the same level of intrinsic merit: a privacy shroud that could be broken by an ambitious elementary school kid.
I, for one, welcome our new mules.
Bringing Gamma Back
The guest, Li-Huei Tsai, director of some august learning and memory institute at MIT, eventually confesses that transfer from mice models to humans in this line or research has about a 1% success rate.
It's a great episode.
Memories retrieved in mutant 'Alzheimer's' mice — 16 March 2016
Proof used to be universal. Now speaking to your base involves one form of "proof", while conversing with educated humans involves another.
Many people don't seem to mind being called base, because it implies staggering levels of power (alongside staggering levels of stupidity). Unfortunately, Donald seems to have lost the tired-of-winning remote engine starter fob under some Whitehouse sofa cushion, since replaced, and now all that staggering power remains parked in the driveway.
Looking stupid.
And not making enough noise to make stupid look awesomely impressive.
Though it sure looks like it could make that much noise, if ever Donald found the keys again.
Wow, IBM attempted to throw more hardware at the problem, and actually succeeded.
Downside: requires more hardware.
It probably does decrease average training latency at scale (not counting the shovel time invested in sunk cost), but I'd still hesitate to call this an advance.
Those 'impressive' Greeks would fall off their pedestals if they had half an inkling of the amount of knowledge the average broadly read and well-informed IT geek of today carries around as a matter of course.
The Library of Alexandria had somewhere between 40,000 and 400,000 scrolls.
I suspect that 10% of these articles would be as large as any dozen ancient scrolls.
I figure the keyboard and the piano will walk into the sunset arm in arm, i.e. not any time soon.
You're dreaming if you don't think you run a similar risk with Intel. The only difference here is the proximal news cycle.
Tomorrow's Market Probably Won't Look Anything Like Today
Well, I suppose there are worse problems in life than paying 20% more for 20% less because of an edge case.
Unless you make it a habit, and it begins to consume larger fish, like your 401(k).
I glanced through Sarah Mei's Twitter page, and she's full of shit. She seems not to get why women in tech might not be evenly distributed.
* Suppose you have a culture that hires based on personal referral. (It's usually one of the best ways to go.)
* Suppose your culture starts out with a male nucleus.
* Suppose your male nucleus mostly has male contacts.
You're gonna get a mainly male culture.
Companies don't hire the best candidate available. Companies hire the candidate for whom they have the most confidence of strong performance, meaning that the route into the door matters a lot. Applicants at large will not be given equal shrift to applicants with a strong, internal referral.
From that starting point, the organization is subject to network effects, none of which need to be intrinsically biased in order to lead to a biased outcome (as determined by simple headcount).
One can argue that the sorry state of women in technology justifies taking active measures against the default behaviour of your (potentially) gender-neutral starting point. One can't argue that failing to take active measures automatically incriminates your starting point as gender discriminatory.
In Sarah's world where water isn't wet, and laudable corporations seek the best candidate while paying no attention to existing network effects, you can draw these conclusions, loudly and with no nuance, should it serve your purpose.
I'm not saying that innocent bias doesn't coexist with toxic bias. I am saying that presumptive guilt is an extremely dangerous tool as wielded by a small, angry imagination.
From The custom TRS-80 & other mysteries"
I found that quickly because I can still recall the entry point 0x1BB3.
I got that wrong in my mind, didn't I? It takes you back to the call point. Oops. Difficult business, this is, dredging up memories from once-upon-a-forever-ago.
I'm pretty sure I wrote that program for maximal obfuscation, so there could have been another for loop in the subroutine, and I really was ping-ponging wildly.
My high school purchased the original 4 kB version, and then upgraded it to 16 kB the following year. The school generously allowed me to take it home on weekends. There were two Apple IIs in the lab as well, with actual floppy disks. I chose the TRS-80 simply because I could get more time. The Apple IIs were busy playing games most of the time. Did anyone learn to actually code on those machines? Not that I noticed.
This was my first actual computer. In my first week of programming, I tried to write:
...
if 0 <= i < 100
I was shocked when this didn't work as expected. Man, I thought to myself, these things are even more brutally literal-minded that I had ever imagined, and I had already been reading about computers for years.
After that, I constantly had at the back of my mind: just how broken is this language?
Well, you can write a for loop. Inside the for loop, you can call a subroutine. Inside the subroutine, you can write next. This takes you back to the for loop. Then inside the for loop, you can write return. This takes you back to the subroutine. Only don't do return again, your call stack is empty. You need to get out again with another next statement. Yes, I actually wrote this program using the TRS-80 BASIC.
It wasn't long before I wrote a skeletal Z80 disassembler in BASIC and then reverse-engineered the ability to load machine language programs directly from cassette tape.
After that, I didn't program in BASIC very much.
I still have a TRS-80 reference manual with the full schematics.
No, my idea doesn't work, because the Mafia can do the same thing in reverse: gather up all the receipts associated with "paid" votes, then randomly test ten (a $10,000 cost-of-doing-business fee), on penalty of worse-than-death.
I think that would reduce the enforcement cost enough to turn paying for votes into a cash-flow-positive business model.
Bear in mind that delivering on penalty of worse-than-death is not cheap (either in time now, or potential for time later). If all the rabbits are trembling enough, you won't need to do this.
With suitable encrypted signatures, you could set up the receipt so that you had to pay a $1000 fee to check that your vote was accurately counted, and if it wasn't, you are awarded $1,000,000 in compensation (to deliver the sting where it matters, each election oversight group could be organized like the unlimited Lloyd's "names" of old).
Why would you do this?
Well, there's a problem with ubiquitous, easily verified receipts that's usually covered in the second lecture of Public Administration 101.
I don't presently feel like Googling that for you.
Have you been sleeping under a log?
The entire brilliant new thing about deep learning is that you can build an entire machine translation system from fucking randomized matrices, all the way up, where no-one got to decide anything.
Hand-crafted rule-based systems present thousands of opportunities for power-mad silverbacks to dicker to their own advantage (see Swamp, The).
But with deep learning, you bootstrap the system with massive artefacts extracted from the real world (the training corpus) and even if you wanted to dicker with the artefact, we've got barely the first clue about how to tilt the artefact—bear in mind that it's very, very big, with a low center of gravity—so that the machine learning algorithms respond in a desired, predictable, stable way (that isn't entirely upended by the next trivial dicker).
Wake up and smell the bacon. Gradient is out there, and mankind no longer sits at the top of a micro-manageable food chain.
This has always been true in small corners of human affairs. At the end of the day, it really doesn't matter who invented the calculus. For the most part, the calculus is a Platonic good.
Networks can lie across a fairly wide swath of the Platonic–political spectrum.
We kind of lucked out with the Internet. ARPANET desired a certain form of resistant against politics (i.e. the backbone coup) that aligned with the individual's preference against being controlled (they didn't foresee, I don't think, how soon social media would become a larger stakeholder than the Pentagon, or more of the ribbon-breasted control freaks would have popped out of the ARPA-oversight woodwork).
If our explicit goal is to tilt way over to the Platonic side, it's not like we have a huge number of dials to bicker about, anyway.
Resource management requires some kind of accounting system which identifies endpoints (bandwidth is neither infinite nor free when push comes to shove). I don't know whether our anonymous micro-currencies are up to the job yet, at such enormous scale.
How does one respond to a DOS attack on a fully onion-routed fabric? Sounds like a tough problem. If it's not onion-routed, there's clearly a small privacy leak that could be exploited by nation-state agencies.
Real problems.
If this ends up becoming a voluntary network (you can join your node if you want to), then like all good libertarian systems, the primary vote is conducted by the pitter-patter of many feet.
In such a world, when the technical committee gathers together, they are going to look around the table to see whether the assembled group has the competence and credibility to prevail in a vote of the feet—because otherwise they're just squandering their time and reputation to get involved in the first place.
So there's you final answer. At the outside boundary condition, we all decide.
Internal to this, Newton will either decide to work alongside Leibniz (good idea if he wishes to succeed) or not. So, yeah, if your amygdala is so inclined, there will likely be a spot of Alpha Geek Mean Girls during the voyage, that you can happily point to forever after as responsible for any lingering imperfections of Internet 2.0.
That's the ultimate in couch-compatible issue trackers: 999 valuable reforms all blocking on "solve human nature". Congratulations, you are now the proud owner of a labyrinthine saddle point that stretches as far as the eye can see in 200 fucking dimensions.
If it weren't for the giant "who decides" monolith erected at consensus centroid of Saddleplane Peak Perplexity, no two people wandering alone in that vast undulating outback would ever meet up to exchange bile.
Well, sure nice to see a human face every so often. Best of luck to you. Me, I'm heading thataway ...
Let's also mention your invocation of "all Christians are created equal".
Publicly they move as a group (sort of, sometimes), privately they bicker lick hell. But they sure love to sell "we are all one" when it allows them to assert their particular brand of moral spray paint.
Catholics and birth control are high on the list of All World hypocrisy gradients.
Let's consider another case. On the plus side, they got Neil Gorsuch. On the down side, they just sold their collective souls to a devil who is a real life cartoon figure possessing real life nuclear weapons, but bereft of any emotional, spiritual, or moral hesitation whatsoever to turn the other ass cheek at the least provocation.
(What would Jesus do? He would run. Run, run, run far away. Nice guy, backbone not included. Think about it. If he is still in the business of carrying people along the sandy beach of life, he lately seems to be pointed in the wrong direction. Hey, Jesus, if you're listening, get yourself a moral compass and turn this damn ship around.)
Meanwhile, the Victorian era just called. They want their table drapes back.
Going ape shit over the FreeBSD logo—to give you troll post any credence at all— is a form of triviality porn, a favoured "gotta do it my way" micro-penis performance art of legacy privilege.
Relative importance of the FreeBSD logo semiotics compared the daily news cycle: about 10^-12.
Finally, I guess I bit the wormy apple after all.
You won. Lucky you.
I would just love to see you in action in a focus group round table discussion.
Why are all these other people here? Trust me, I've got it covered.
Most of the comments I skimmed are missing the point.
The real problem is that even scientists with the best training and the best intentions wind up committing a certain amount of p-hacking subconsciously. Just a simple data exploration to decide post hoc whether any collected data is corrupted or implausible, and you've already slithered one toe across the p-hacking line.
When p-values gate publication, and publication gates promotion, you create a severe moral hazard where many of the scientists you end up promoting lie on the bottom half of the curve in self-policing their accidental p-hacking. The guy with the penchant to do slightly more irregular experiments, which require slightly more data cleanup, seems to get slightly more published results. Ba da boom.
p=0.005 would put a pretty big crimp in this effect.
Of course it doesn't solve the larger problem. But good golly, first things first.
We also know from replication efforts that p=0.05 is allowing far too much crap to float over the gate. p=0.005 probably gets us closer the crap level we naively assumed we'd get when we originally rallied around p=0.05.
Probably the increased use of computers hasn't helped matters: even accidental p-hacking with pencil and paper is hard work.
The purpose of my post was to troll-shame the one-trick ponies, which was so easily accomplished, I managed to press "submit" without even stopping to think.
If teams with women have a different pattern of returning hat-in-hand for a subsequent round (perhaps with better foresight and anticipation, hence less desperation and better leverage) then the VC could very much end up munched in the ass by imprudent gender equality.
The correct metric here is return to VC investor. Perhaps teams with women are less easily shaved in a subsequent down round. Then it would be rationale for VC investors to eschew EQ on the business-end of the puppet strings.
Missed a trick, being too accustomed to almost making myself comprehensible.
I'm so ashamed. It would have served as the perfect foil for my subsequent riff about withholding the verb for too long.
Actually, the problem here is that the context is cold. It would be perfect in this more extreme form after warming the reader up with some sentences thoroughly stilted in the other direction. But in this piece, this sentence is positioned in the transitional foyer, and the extreme form is too abrupt.
On second glance, my shame decays into a shrug.
With this kind of humour (supposing there is any) it's often useful to ask yourself along the way, "well, who would be the perfect reader for this piece?"—because if you can't name even one perfect reader, it's likely that you have phased entirely out of the plain of comprehensibility.
Scala-loving Dinesh Chugtai would be a good choice. He'd definitely have laughed at my turn of phrase "Pandit Chin Thumb". And he'd also at least detect my scatological / linguistic / functional programming triple-entendre at the very end.
What does dot colon colon (.::) mean in Scala and why does it remind me of colon slash slash slash dot?
Unfortunately, my prospective perfect reader is also himself a satirical figment.
Danger, Will Robinson!
What follows is a meteor shower of post-historical Chicago Manual compound-modifier satirical punctuation.
Configure ears in the upright, locked position, and proceed on impulse power only.
___
I'm pretty sure a fullname–verb ndash is properly ASCII-rendered as Norman_Foster-designed.
This could have been one of the best double-pips ever, if our submitter were irony-enabled about Slashdot's Flintstones–Jetson retro-chic electronic heritage.
Then the headline could have been: How Jony's Norman_Foster-inspired monument to Pandit Chin Thumb ransacked a Death Star fashion parade.
(Unfortunately, my sense of humour often makes the highly mistakable[*] blink--and--you--miss-it-twice "whoosh, whoosh" sound of a side-holster unfriendly dipole photon saber.)
[*] Do notice my clever "-ly" guru override.
Rewriting the full sentence to also address the weird colon:
This sentence, as properly recast with the long, orthodox, purely phrasal appositive, also has its own, internal humour, with an olllllllde-English Teutonic tonsil-vibe inherent in the long-withheld sentence-final verb.
Come to think of it, too much withholding the verb was the subject of a bitter bear-cam quarrel in The Wolf of Wall Street.
Perhaps the colon is well suited to this innovative folding-function, after all.
I'm not trying to be difficult or argumentative, but Grandma always seemed like such a useless person.
Yeah, you're right, but she doesn't take up much space and she's the only person in the house who can consistently walk into a strange kitchen and crank out perfect fettuccine noodles without having to hoist an unfamiliar stand mixer onto a foreign countertop, and then dig through all the cupboards for a noodle attachment, if the cheap IT bastards even have one lying around.
But she's really old and out of date and doesn't even use ribbons!
Well, suit yourself, but I'd rather keep the old bird around, in the handy drawer in the closet under the staircase, where she's been living quietly without making a fuss for the last 35 years.
Yes, so long as you aren't simultaneously sustaining any other thing. Like a day job.
I'm joking just a bit, but the word "sustain" is commonly abused in exactly this way.
Weakly sustainable: when just this one thing can be sustained.
Strongly sustainable: a member of the set such that all strongly sustainable things can be sustained at the same time without surpassing the labours of Hercules.
Whenever someone says to me "sustainable" regarding a personal resolution, my first (usually silent) question is: have you ever given one hour notice at work, and then set foot in Tibet the very next day?
Because, if so, that's just a steaming pile of dedication porn.
His satisfaction is quite irrelevant. Unless you believe that Any True Scotsman would faff around indefinitely to shave one more nickle off the purchase price.
What matters here is his prospective utility: his net upside after the huge investment to research the alternatives, reinvent his established workflow, learn about all the new nits and gremlins, flawlessly administrate his custom stack of validated alternatives, resolving interoperability difficulties with his contacts and clients, etc. etc.
About 10% of all open source zealots really ought to check themselves into Faffaholics Anonymous.
The thing is, you sign up as a Republican, and on the way to work, the company who employs you suddenly turns into Trump University, and then one day, like Morning Joe, you announce to the public: "you know what, I'm just not into working for Trump University even though I'm going to continue to be a Conservative".
Now, it is true that a blind partisan Democrat can be assumed to still be a Democrat ten years later (even if they did nominate loathsome Hilary), but do we really know the guy ticked off "blind" and "partisan" and "until death do us part" on his original Democrat Vow of Perpetual Allegiance?
I'm sure there are many Obama Democrats who signed up because of Obama, and then checked out of politics during the last election cycle when the Dems nominated a previously-owned, pant-suited, Wall St toady (if they hadn't already checked out halfway through Obama's first term, when he proved to be a colossal disappointment in standing up to the financial sector).
Furthermore, we have strong evidence of the great Democrat check-out in the form of Trump's ghastly victory. We're not even sure he could have won this election running against a ham sandwich, although he did win this election running against "crooked" Hillary. Even though Trump personally thinks Hillary is infinitely worse than a ham sandwich, he still finds time to brag about his victory as a meaningful achievement.
Hillary: high bar or low bar? Pick one.
For myself, the most effective thing I could do politically would be to register with my regional party on the right (I'm not American) so I could help nip against the darling single-issue candidates of the Christian Right in the bud at the nomination stage. This affiliation wouldn't be too hard for me to pull off, because I actually believe in the good half of Libertarianism—compassionate Libertarian wouldn't even be that far from the truth (though my opportunities to fully align myself are thin on the ground).
Then I would be, by American standards, a registered Republican. I'd be happy enough and it would all go swimmingly until Scaramucci shows up, and only then, I would totally blow my cover.
Anthony Scaramucci in 2010: Wall Street feels like a 'pinata'
There are hapless Greenpeacers on the left who say equally ridiculous things (splitting desk with head ridiculous), but they're into Volvos and macrame and other mostly harmless things, whereas Scaramucci is a spoiled child of privilege, and I just can't stand fucktards who defend their ridiculous views by pointing at their bank accounts—in a disgusting show-boat of glitzy ad hominem argument (the good half of Libertarianism believes that all men are created equal, the bad half of Libertarianism believes that all dollars are created equal; a die-hard Conservative is someone who conveniently neglects the difference when ideology hits the ballot box).
People are complex.
America's hidden philosophy — 18 July 2017
I didn't like this article much, but it did make one or two good points. Rational choice theory has long aligned itself with reductive analysis.