Eric S. Raymond's Blog, page 25

Reposurgeon against the Martians!

I released reposurgeon 3.30 today. It has been five years and a month since the first public release.

In those five years, the design concept seems to have proved out very well, finding use in many repository conversions. But the project exhibits an unusual sociology; I don’t get lots of casual contributors, only a few exceptional ones.

Your typical open-source project sees a sort of exponential distribution in which small fix patches from people you see stop by only once are common, single feature-sized patches less so, and complex sustained work that reimagines entire subsystems is rare. There’s an obvious inverse relation between frequency and complexity scale. At intermediate and higher complexity scales you often get regular contributors who do extended work on different things over time. GPSD is like this.

On reposurgeon I see an entirely different pattern. Casual patches are rare to nonexistent. For long stretches of time I have no active collaborators at all. Then a hacker will appear out of the void and begin contributing very clever patches. He (no shes yet) will draw closer to the project, and for a few days or weeks we’ll be in an intense collaborative mode tossing ideas and patches back and forth. Some complex series of features will be implemented.

Then, his particular feature-lust fulfilled, said hacker will quietly vanish into the interstellar darkness never to be seen again, like some comet on a hyperbolic trajectory after a pass near the Sun. Never yet has there been more than one cometary hacker at a time.

OK, I exaggerate slightly. The project has some semi-regular hangers-on in the #reposurgeon channel (one of them is A&D commenter Mike Swanson). But those people tend to be power users rather than major code contributors; the pattern of large code drops by people who appear, do work that impresses the hell out of me, and then vanish, still dominates code contributions.

My wife Cathy called this one right when I remarked on it. Most people never use reposurgeon more than once, but the hands that find it are disproportionately likely to be very skilled ones. All of my half-dozen or so cometary contributors have been damn good hackers even by my elevated standards, careful and imaginative and tasteful. When people like this detect a deficiency in a tool, they fix it – and their idea of “easy” fixes would daunt lesser mortals.

It’d be nice if some of these hackers would stick around, because I love collaborating with people that bright, but oh well. They’re as in demand as only the capable can be. And at the end of the day, there are much worse things you can say of a software project than “it attracts high-quality work from high-quality people, er, even if they don’t tend to stick around”.

Why I joined the NRA

After 20 years of evading joining the NRA, I finally did it last week.

I’ve never been a huge fan of the NRA because, despite the fearsome extremist image the mainstream media tries to hang on it, the NRA is actually rather squishy about gun rights. A major symptom of this is its lack of interest in pursuing Second Amendment court cases. Alan Gura, the civil-rights warrior who fought Heller vs. DC and several other key cases to a successful conclusion, was funded not by the NRA but by the Second Amendment Foundation. Also, in the past, the NRA has been too willing to acquiesce to unconstitutional legislation like the 1986 ban on sales of new automatic weapons to civilians.

So, you might well ask: why am I joining an organization I’m dubious about now, when the gun-rights cause seems to be winning? Popular support for Second Amendment rights is at record highs in the polls, a record seven states now have constitutional carry (no permit requirement), Texas just became the 45th state to legalize open carry last week…why am I joining an organization I’ve characterized as squishy?

I joined because the state-worshiping thugs on the other side are doubling down, and they still own most of the media and the machinery of the Federal government. After decades of pretending that they only wanted soi-disant “common-sense” legislation aimed at specific problems around the edges of gun policy, the Democratic Party is now openly talking of outright gun confiscation. The usual suspects in the national press are obediently amplifying their propaganda.

Some things you do for substantive effect – giving money to the SAF so Alan Gura can win another case is like that. Some things you do less for effect than as as a signal of pushback intended to create political momentum and demoralize the other side; joining the NRA is like that.

Meanwhile, I think my sentimental favorite gun-rights organization is still Jews for the Preservation of Firearms Ownership. Because they have the motto that truly says it all to those with any sense of history: “Never again”.

Beehive huts to the stars!

None of the things I expected from seeing The Force Awakens was to recognize the location where the last scene was filmed, because I’ve been there myself.

I’m being careful not to utter spoilers here…but I’ve been to the western coast of Ireland, in County Kerry, in a place called Fahan. And in that place, where the Atlantic crashes on the shore and blue of the sky and the green of the grass are more vivid than anywhere else I’ve seen on Earth, there are beehive-shaped stone huts called “clocháns” on the hills that tumble down to the sea.

There are many legends about the clocháns, the most picturesque of which is that they were built by hermit monks around 1000CE. Built by hand, without mortar, they are rude and dramatic ornaments to a landscape that would be pretty impressive even without them.

The combination is unmistakable, and that’s how I know pretty much exactly here that last scene was filmed, to within a couple of hundred feet, on the south-facing shore of the Dingle Peninsula. I’ve seen it with my own eyes; I may have walked on the same paths the movie characters used, probably did in fact.

And again, no spoilers, but…it was a superb choice of location for that scene. Well done!

(And for you smartasses out there, no it wasn’t the huts on Skellig Michael rather than the mainland. The lie of the slope was wrong for that; besides, schlepping a film unit to the island would have been both hideously difficult and pointless when the shore locations were about as good for what they wanted.)

UPDATE: On new evidence, they changed locations during the scene – I was right about the beginning, but the very last bit (like, the last 60 seconds of the movie) was indeed filmed on Skellig Michael.

Comparative language difficulty for English speakers

This morning I found a copy of the chart the Foreign Services Institute uses to grade the comparative difficulty of world languages for acquisition by an adult monoglot English speaker.

I have an unusual perspective on this list for an American. I’m a low-grade polyglot; I have spoken three languages other than my birth English and can read a couple others with Google Translate. I have studied comparative linguistics; I know a bit about the morphology and phonology of many of these languages. I have received street-level exposure to over a dozen of them in my extensive travels, and I have a good ear.

So, I’m going to try to add some value to the list with additional notes and comments.

First, the list itself:

Category I: 23-24 weeks (575-600 hours)
Languages closely related to English



Afrikaans

Danish

Dutch

French

Italian

Norwegian

Portuguese

Romanian

Spanish

Swedish



Category II: 30 weeks (750 hours)

Languages similar to English


German




Category III: 36 weeks (900 hours)

Languages with linguistic and/or cultural differences from English



Indonesian

Malaysian
Swahili



Category IV: 44 weeks (1100 hours)

Languages with significant linguistic and/or cultural differences from English



Albanian

Amharic

Armenian

Azerbaijani

Bengali

Bosnian

Bulgarian

Burmese

Croatian

Czech

*Estonian

*Finnish

*Georgian

Greek

Hebrew

Hindi

*Hungarian

Icelandic

Khmer

Lao

Latvian

Lithuanian

Macedonian

*Mongolian

Nepali

Pashto

Persian (Dari, Farsi, Tajik)

Polish

Russian

Serbian

Sinhala

Slovak

Slovenian

Tagalog

*Thai

Turkish

Ukrainian

Urdu

Uzbek

*Vietnamese

Xhosa

Zulu



Category V: 88 weeks (2200 hours)

Languages which are exceptionally difficult for native English speakers


Arabic

Cantonese (Chinese)

Mandarin (Chinese)
*Japanese

Korean


* Languages preceded by asterisks are usually more difficult for native English speakers to learn than other languages in the same category.

Generally I think this list is dead on target. I agree with most of it even to the level of which languages should be starred in their category.

Before I launch into specific discussion of the exceptions, a minor caveat that I think I might have a tendency to underweight phonetic difficulty because of my Frodo ear – there are very few phonemes that really throw me for a loop, and I can hear tones with little effort.

That said, here we go:

Arabic should be starred. It is grammatically and phonologically extremely difficult; from my exposure to both I’d say significantly more difficult than Chinese without traditional Chinese writing.

If I were to add a mark for “easy in its category”, I would put that on Spanish and Italian. These are significantly easier than the other Category I languages and (I believe) the easiest of all world languages for English speakers to learn.

Portuguese is a freebie if you learn Spanish and can get used to an odd but relatively consistent shift in the phonology, mainly heavy nasalization of everything.

I think putting German in an intermediate category II above other major European languages in difficulty is very shrewd. Phonology is easy (I can pronounce German with a good accent even though I don’t speak it) but the grammar is significantly more difficult than say, French.

If not for the help German gets from having Indo-European cognates, I think Category III (Indonesian/Malay and Swahili) might be easier than German. These are areal trade languages, quite probably creolized from ancestral trade pidgins, and have retained a simplicity that makes them relatively easily acquired by adult speakers. (English has a similar history.)

Chinese is a strange case. It is indeed brutally difficult in toto, but if you (a) are willing to settle for spoken fluency only, and (b) have a musician’s ear for tone (as I do) I suspect it falls to category IV and possibly to a hard category III. The positional grammar of Chinese is simple and easy to acquire.

Given an “easy” mark I would also rate Polish an easy class IV for an educated English speaker, mainly due to a much heavier infiltration of Latin roots than in other Slavic languages – this make recognizing cognates easier.

Persian/Farsi/Dari rates an “easy” mark in its category due to simple and regular grammar. If I were going to try to grok the Indo/Persian group I think it’s a tossup whether Farsi or Hindi would be the best point of entry.

Lithuanian deserves a star, I think. Phonology isn’t bad but the grammar will break your brain.

Xhosa definitely deserves a star. When the phonology is difficult enough that I have trouble retaining the distinctions, most English-speakers would be lost beyond hope. Clicks and implosives, man!

On the blog where I found this list, a commenter opined that there ought to be a category zero for Esperanto. Yep.

A short course in counter-terror theory

In the wake of the San Bernardino shootings, more Americans than before are trying to grapple with questions about the nature of terrorism, terror activity versus rampage killings, and what can be done to prevent these bloodlettings.

I have been studying these questions for years as part of my self-training. I learned some of the basics of counter-terrorism theory from a former SpecOps officer, and more from my Kung Fu instructor whose day job is as a criminal forensics and counter-terrorism specialist consulting to law enforcement. I’ve also read up on the subject, and thought carefully about what I’ve read.

The following is a primer on how people whose job it is to prevent and mitigate terrorist activity and spree killings think about it.

First I’m going to present a couple of perpetrator types which, between them, account for almost all terrorist acts and rampage killings. Later I’ll point at some edge cases and exceptions.

Our first perpetrator type is what I’ll call a “terror soldier”. This is not a term in use among professionals, I’m presenting it here in order to avoid pre-empting the term “terrorist”.

The terror soldier does not act alone. He has a network behind him; the network provides him with, at minimum, ideological and tactical direction. It may also provide him with safehouses, money, and weapons. Because the terror network has public political objectives, it either has an above-ground political arm or a deniable conduit to a “legitimate” political organization that can operate as its propaganda and recruiting arm.

The opposite archetype is the lone wacko. The lone wacko doesn’t have a network. His motivations are not public and political but personal and, usually, delusional. He is likely to have been a former mental patient, or to have a history that clearly indicates previously undiagnosed mental illness when it is scrutinized after he has gone violent. More often than nor he will have been on long term use of SSRIs (more than five years) or some other prescription antidepressent or antipsychotic medication, and the violent break will be associated with going off or changing his meds.

An important contrasting point about terror soldiers is that they are usually not clinically nuts. Terror networks (like covert operations in general) avoid recruiting mental cases because they’re brittle and unreliable.

A terror soldier has signed up for a bit part in a political war planned by others. A lone wacko is his own reason for slaughter; his motivations are not political but psychological and often inscrutable to anyone outside of his skull.

Generally, when you hear the phrase “terrorism” used to describe a mass shooting or bombing, terror soldiers are at the back of it. Generally, if you hear of a “spree killing” or “rampage killing”, a lone wacko did it. There are some exceptions, which I’ll get to.

These distinctions are important because lone wackos and terror soldiers have different threat profiles. A lone wacko is a tragedy but almost never a disaster; a terror network can scale up violence to much greater levels by deploying multiple soldiers, and is far more likely to have expertise in bomb-making, airplane hijackings, and other means that can inflict casualties well above the level of a rampage shooting with personal firearms.

The difference even has tactical consequences; armed civilian shooters are highly effective at stopping lone wackos, and a major reason is that wackos tend to freeze and surrender when shot back at. This is not the case with terror soldiers.

Thus, if you are a pro, you worry a lot about terror soldiers and much less about lone wackos. The lone wackos are much less dangerous, and (as we’ll see) there’s less you can do against them anyway.

These distinctions are also important because lone wackos and terror soldiers require different counterstrategies. To stop terrorists, you penetrate and destroy their support networks. In fact, you must do this, because unless the network is taken out it will recruit and indoctrinate more soldiers.

Lone wackos can’t be attacked through their networks because they don’t have any. The only way you could stop them is by massively increasing screening for mental illness and getting tough about involuntarily committing the potentially dangerous. In the U.S. the political will to do this has been absent since the de-institionalization movement of the 1960s; attempts to reverse that would run into both cost problems and serious civil-liberties objections.

The San Bernardino shooters were terror soldiers, not lone wackos. There was more confusion about this than there should have been until Tashfeen Malik’s declaration of allegience to ISIS was discovered. Some people are still confused by the fact that ISIS didn’t provide them material support. But this is exactly what makes ISIS novel and dangerous – it has built a doctrine and toolkit for running soldiers with ideological and tactical direction only, purely through its propaganda arm.

By contrast, the Aurora Theater shooter, James Holmes, was a classic lone wacko. So was Elliot Rodger, the shooter in the 2014 Isla Vista killings. So were Harris and Klebold, the Columbine High School shooters. No network, no ideology, just boiling cauldrons of private hatreds and resentments.

Almost all terrorists or rampage killers fit one of these two profiles. Very occasionally you get some outliers that break the classification. Timothy McVeigh was one – politically motivated, military demolition skills, not mentally unbalanced: every attribute of a terror soldier except the network.

The Unabomber was similar, except borderline crazy. Unlike McVeigh he might not actually be an exception to the usual rules but, rather, best understood as an exceptionally intelligent lone wacko. In the real world, you can never count on category boundaries being perfectly sharp.

One recent borderline case deserves special attention: Dylann Roof, the virulent racist who shot 9 people in a Charleston church in June 2015. Dylann is best understood as a would-be terror soldier who, in contrast to the self-radicalized San Bernardino shooters, failed to find a network to hook up to. There was no ISIS for him; his Facebook stream included complaints that he couldn’t find any racists to hang with.

But I emphasize that Roof and McVeigh and the Unabomber were exceptions; 99% of the time it is very obvious that you are dealing with either a terror soldier (backed by a network) or a lone wacko from just the modus operandi of the killing, and there is almost never any need to change this assessment later.

I’ve been talking as though terror soldiers and lone wackos have roughly the same prevalence, but in fact lone wackos are both less dangerous and far, far more common than terror soldiers – the difference, in the U.S., seems to be roughly an order of magnitude. This has an important consequence; if you’re even a little bit off about distinguishing them, your threat model of the most dangerous bad guys (the terror soldiers) will be badly compromised.

Now I’ll get to the controversial part. A lot of the confusion about terror and rampage killings is politically generated and unnecessary. The operators on the ground are seldom in much doubt about what species they are dealing with, but they’re used to seeing their analyses spun into garble, vagueness, and sometimes outright fabrication by their superiors and the news media.

There are two major reasons for this. One is that for PR reasons, the U.S, government has chosen to underplay the role of Islamist indoctrination in recent terror incidents, implicitly binning terror soldiers as lone wackos. Perhaps the single most egregious example of this was in the 2009 Fort Hood shootings, which the government insisted on publicly categorizing as “workplace violence” despite the fact that the shooter screamed “Allahu akbar!” while firing and the Joint Terrorism Task Force found him to have been communicating with a jihadist imam in Yemen. Co-workers had been aware of the shooter’s increasing radicalization for years.

Another major reason is that the left end of the American political spectrum is heavily invested in the belief that “right-wing terrorism” is prevalent in the U.S. and a greater danger than either left terrorism or Islamism.

This belief is a myth. One recent indicator is the fact that Dylann Roof, a natural hard-right-wing terrorist soldier if there ever was one, never found his network. Another is that anybody can name Islamist terror organizations that operate in the U.S. – ISIS, al-Qaeda – but only specialists know about U.S. incubator networks like The Order and the Christian Identity movement.

In fact, the potentially-terrorist hard right in the U.S. is tiny, isolated, and so incompetent that it can barely find its own ass with both hands, a flashlight, and GPS guidance. It is also heavily infiltrated. (This is not just my opinion, it is what any pro in the field will tell you if you can get them to talk.)

What sustains the myth that right-wing terror is more prevalent than jihadism is, basically, the news media instantly counting any lone wacko with a white skin as a “right-wing terrorist” and sticking to that categorization even when facts contradict it.

This bias is so extreme that Joseph Stack, who flew a light plane into an IRS office in 2010, is still routinely described as “right-wing” even though his suicide note ended by quoting the Communist Manifesto! Another notable example is Jared Lee Loughner, characterized as “right wing” even though his political connections were an incoherent mess of mainly left-wing conspiracy theories and a former classmate testified that he was “left wing, quite liberal” before retreating into private psychosis.

I have to single out for particular opprobrium the Southern Poverty Law Center, an organization that used to do noble work in civil rights but has in recent years been particularly persistent and dishonest in promoting the myth of pervasive right-wing terror. Journalists still treat them as a reliable source, and should not.

Another political hobbyhorse that gets ridden after every mass shooting, whether terrorist incident or rampage killing, is gun control. This article is not really the place to fully analyze the kind of dimwitted magical thinking involved, but I will note one relevant fact: both terror soldiers and rampage killers are known to preferentially seek out posted “gun-free zones” – venues where they are reasonably confident their victims will not be able to shoot back.

Turning away from the politics, ignoring Islamist connections and mis-classifying lone wackos as right-wing terror soldiers are bad things because they distort threat modeling and countermeasures. Reality is what it is. To hold terror and rampage-killing casualties to a minimum you have to face that reality rather than substituting a politically convenient narrative.

A rational strategy for addressing mass killings in the U.S. would include the following elements:

1. Traffic analysis of social-media posts. Terror soldiers leave one distinctive kind of trail, linking to their network. Lone wackos leave another – they tend to be socially isolated by mental illness.

2. Stepped-up efforts to identify violently unstable people and institutionalize them before they flip out. An end to over-prescription of SSRIs and other medications that screw with neurotransmitter balance.

3. More armed civilians and the elimination of “gun-free zones”. (This has been tested and found effective in Israel. In the wake of San Bernardino we are beginning to see police organizations recommend it.)

4. Conventional-force expeditions to recapture or destroy the home grounds of organizations like ISIS and Boko Haram.

These show up repeatedly in every conversation I’ve had with anyone knowledgeable in the field. There may be be good political or ethical reasons for not doing some of them (I’m rather wary of reinstitutionalization, myself), but we must acknowledge that such high-minded restraint will have a cost measured in lost lives.

I will also note some measures which might be effective but are debatable among professionals:

* Tighter border controls. They may work tolerably well in a small, ethnically homogenous country; it is unclear that these can be made long-term effective enough in the U.S. We have porous borders just as a consequence of geography and economics, and a multiethnic population that infiltrators can easily disappear into. On the other hand, not actively pulling in refugees from Syria or other majority-Islamic countries is probably good short-term policy, and there are other good reasons (like the child-rape rates in Rotherham and Malmo) to cut that flow as much as we can.

* Encryption back doors and Patriot-Act-style mass surveillance of content. There are are hot disputes about this. My own judgment is that they are a bad trade-off, or at least that if they are effective the pro-surveillance advocates have dismally failed to produce evidence of same. On the other hand, there are imaginable good reasons for the evidence itself to be suppressed.

I don’t have any neat conclusion to wrap this up with. Or, wait, maybe I do. There’s one other common, major mistake: treating terrorism and rampage killings as crime problems. Neither of them really is.

Rampage killings are a public health problem – police may be the first responders to an incident, but the effective interventions to prevent them them are mainly medical, not criminological.

On the other hand, what terror soldiers do is best thought of as a kind of distributed irregular warfare, intended like all warfare to break the enemy’s will to resist. Criminal enforcement can typically do little or nothing about their networks. Instead, the normal counter to irregular warfare applies; you want to bait them into concentrating so they can be confronted and destroyed by regular forces.

In that respect, ISIS presents an easier target than al-Qaeda did. By proclaiming a caliphate and holding ground, they’ve fixed themselves. It remains to be seen if the West has the political will to actually confront them.

On the shooting of Laquan McDonald

One of my regulars, mindful of the forensic analysis I did on the Michael Brown autopsy photos last year, has asked me to comment on the Laquan McDonald shooting from my point of view as a pistol and self-defense instructor.

The fast version: I would have said this was what cops call a “good shoot” if it had stopped at the first two bullets. It didn’t. I don’t think this was murder one, but it was at least criminally negligent homicide and those who covered it up should be prosecuted along with Van Dyke.

The key portion of the video starts at about 5:19. The blade is visible in McDonald’s right hand; he draws it and brandishes it at 5:25 while facing slightly to the right of a police car that has him in its headlights. At 5:30 you can see that an officer has lined up a pistol on him.

At 5:32 he begins to turn towards the officers. One shoots immediately; he spins and goes down. At that point the officers go out of frame, but we can see at least one dust puff from an incoming bullet at 5:35. We see him either trying to get up off the ground at 5:36 or having a convulsion that simulates the motion; his head and shoulders rise slightly. As late as 5:38 his hands seem to be still moving.

We know from the autopsy that two bullets hit him when he was up and another 14 when he was down (or 15; accounts are inconsistent, and some may be counting at least one round that clearly missed and caused the dust puff).

Now let’s consider this from the responder’s point of view.

The first thing to be clear on is that McDonald was behaving in a crazily aggressive way when he died. You don’t pull a knife and brandish it in the presence of two cop cars if you’re thinking at all sanely.

If I had been a cop on the scene I would immediately have thought “angel dust”, and in fact the autopsy revealed that McDonald was high on PCP. This drug induces violence, freak strength, and insensitivity to pain.

UPDATE: I should have been more specific about the tells here. You can see even in the poor-quality video that McDonald makes a big, rather jerky motion with the knife. A cop (or me) sees that, thinks “Impaired fine-motor control. Oh, shit.” Because at that point the odds on McDonald being fucked up on something like meth or bath salts or PCP rises to the point where you’d damn well better assume it in your planning.

(If you care about fine distinctions, urban black lowlife probably means PCP or bath salts. Rural white trash means meth.)

This is a situation that amply justifies drawing a weapon and preparing to shoot. From the video, McDonald was well inside the 21-foot close-engagement limit – he could have rushed an officer with that knife before the officer could draw on him and trust me that this is not a chance to take with someone you suspect might be on PCP.

If you are any of the cops you are going to be adrenaline-dumping by now. This is a dangerous situation even with your gun drawn; the thug could charge you, take several bullets and still stab you fatally before he goes down. It’s happened often enough before.

Now, he angles slightly away from the group of cops, but they have to be thinking that if he shows any sign of charging they must shoot before he kills them.

I want to impress on my readers that this was a completely justified reaction. Everything the police have visibly done up to this point is textbook procedure for this situation, including what happens next: he turns towards them and Van Dyke, the cop now charged with murder, shoots.

We are still in unquestionable legal and ethical territory until McDonald goes down. What the police have done so far – those first two bullets – is correct.

The next correct thing to do would have been to stop firing for long enough to assess whether McDonald was still a threat. One way this could have gone is: Van Dyke stops shooting, McDonald levers himself off the ground, Van Dyke resumes shooting until McDonald is down again. That would still have been a “good shoot” for which Van Dyke would be neither legally nor morally culpable.

But that does not appear to be what happened. It appears that Van Dyke kept firing continuously at McDonald on the ground. The police report avers that another officer stopped him from firing still more bullets after the 16 he put into McDonald’s semi-fetally-curled form.

It is not difficult to form a plausible theory of why Van Dyke kept firing. As a student of defensive violence I can tell you that under stress this kind of reaction is very common. He may simply have not registered that McDonald was no longer moving.

Matters are complicated by the fact that McDonald may in fact have been still trying to get off the ground and charge Van Dyke, even with several bullets in him. The video evidence is ambiguous on this point.

But even if so, Van Dyke was doing the wrong thing. What he should have done was stopped to assess, realized that even if McDonald was trying to get to his feet that was not going to happen fast enough to put anyone in imminent danger, and stopped firing unless and until McDonald again became immediately dangerous.

Lethal force is a terrible tool. People who use it, whether cops or civilians, must show restraint and good judgment. Van Dyke was, at best, lethally careless.

On the plain evidence of this video, what we have here is a criminally negligent homicide; manslaughter or possibly second-degree murder.

And if it is true that other cops conspired to cover it up, they should be prosecuted too. I can understand their reasoning – why let a cop who made a simple mistake under stress be ruined by the death of a drug-addict lowlife with a knife in his hand? But it was still wrong, because that habit of blue omerta covers up too much.

I do not think the charge of first-degree murder is justified. There is no evidence of premeditation, or reason to suspect it, here. It is certainly possible that the prosecutors know something that I don’t, but I suspect that the escalation of the charge is a purely political maneuver intended to appease those who have put a racial spin on this incident.

And that racial spin? Plain bullshit. Those cops were facing an angel-dusted thug brandishing a weapon; that was pretty much bound to end badly whether the thug was black, white, or purple polka-dotted. But the obsessive scab-pickers of our racial-grievance industry will doubtless attempt to to incite riots over this, and given the media’s usual enthusiastic help they might well succeed.

NTPsec’s beta is released

You’ve heard me uttering teasers about it for months. Now it’s here. The repository is available for cloning; we’re shipping the 0.9.0 beta of NTPsec. You can browse the web pages or clone the git repository by one of several methods. You can “wget https://github.com/NTPsec/ntpsec/arch...” to get a tarball.

This is an initial beta and has some rough edges, mostly due to the rather traumatic (but utterly necessary) replacement of the autoconf build system. Also, our range of ports is still narrow; if you’re on anything but Linux or a recent FreeBSD the build may not work for you yet. These things will be fixed.

However, the core function – syncing your clock via NTP – is solid, and using 0.9.0 for production might be judged a bit adventurous but wouldn’t be crazy. The next few beta releases will rapidly get more polished. Expect them to come quickly, like within weeks.

Most of the changes are under the hood and not user-visible. A few auxiliary tools have been renamed, most notably sntp to ntpdig. If you read documentation, you will notice that what’s there has been massively revised and improved.

The most important change you can’t see is that the code has been very seriously security-hardened, not only by plugging all publicly disclosed holes but by internal preventive measures to close off entire classes of vulnerabilities (by, for example, replacing all function calls that can produce buffer overruns with memory-safe equivalents.)

We’ve already established good relations with security-research and InfoSec communities. Near-future releases will include security fixes currently under embargo.

Why Hackers Must Eject the SJWs

The hacker culture, and STEM in general, are under ideological attack. Recently I blogged a safety warning that according to a source I consider reliable, a “women in tech” pressure group has made multiple efforts to set Linus Torvalds up for a sexual assault accusation. I interpreted this as an attemp to bear the hacker culture into political pliability, and advised anyone in a leadership position to beware of similar attempts.

Now comes Roberto Rosario of the Django Software Foundation. Django is a web development famework that is a flourishing and well-respected part of the ecology around the of the Python language. On October 29th 2015 he reported that someone posting as ‘djangoconcardiff’ opened an issue against pull request #176 on ‘awesome-django’, addressing it to Rosario. This was the first paragraph.

Hi

great project!! I have one observation and a suggestion. I noticed that you have rejected some pull requests to add some good django libraries and that the people submitting thsoe pull requests are POCs (People of Colour). As a suggestion I recommend adopting the Contributor Code of Conduct (http://contributor-covenant.org) to ensure everyone’s contributions are accepted regarless [sic] of their sex, sexual orientation, skin color, religion, height, place of origin, etc. etc. etc. As a white straight male and lead of this trending repository, your adoption of this Code of Conduct will send a loud and clear message that inclusion is a primary objective of the Django community and of the software development community in general. D.

Conversation on that issue is preserved in the Twitter link above, but the issue itself in GitHub has apparently been deleted in its totality. Normally, only GitHub staff can do this. A copy is preserved here.

It is unknown who was speaking as ‘djangoconcardiff’, and that login has now been deleted, like the GitHub issue. (DjangoCon Europe 2015 was this past May/June in Cardiff.)

The slippery, Newspeak-like quality of djangoconcardiff’s “suggestion” makes it hard to pin down from the text itself whether he/she is merely stumping for inclusiveness or insinuating that rejection of pull requests by “persons of color” is itself evidence of racism and thoughtcrime.

But, if you think you’re reading that ‘djangoconcardiff’ considers acceptance of pull requests putatively from “persons of color” to be politically mandatory, a look at the Contributor Covenant he/she advocates will do nothing to dissuade you. Paragraph 2 denounces the “pervasive cult of meritocracy”.

It is clear that djangoconcardiff and the author of the Covenant (self-described transgender feminist Coraline Ada Ehmke) want to replace the “cult of meritocracy” with something else. And equally clear that what they want to replace it with is racial and sexual identity politics.

Rosario tagged his Twitter report “Social Justice in action!” He knows who these people are: SJWs, “Social Justice Warriors”. And, unless you have been living under a rock, so do you. These are the people – the political and doctrinal tendency, united if in no other way by an elaborate shared jargon and a seething hatred of djangoconcardiff’s “white straight male”, who recently hounded Nobel laureate Tim Hunt out of his job with a fraudulent accusation of sexist remarks.

I’m not going to analyze SJW ideology here except to point out, again, why the hacker culture must consider anyone who holds it an enemy. This is because we must be a cult of meritocracy. We must constantly demand merit – performance, intelligence, dedication, and technical excellence – of ourselves and each other.

Now that the Internet – the hacker culture’s creation! – is everywhere, and civilization is increasingly software-dependent, we have a duty, the duty I wrote about in Holding Up The Sky. The invisible gears have to turn. The shared software infrastructure of civilization has to work, or economies will seize up and people will die. And for large sections of that infrastructure, it’s on us – us! – to keep it working. Because nobody else is going to step up.

We dare not give less than our best. If we fall away from meritocracy – if we allow the SJWs to remake us as they wish, into a hell-pit of competitive grievance-mongering and political favoritism for the designated victim group of the week – we will betray not only what is best in our own traditions but the entire civilization that we serve.

This isn’t about women in tech, or minorities in tech, or gays in tech. The hacker culture’s norm about inclusion is clear: anybody who can pull the freight is welcome, and twitching about things like skin color or shape of genitalia or what thing you like to stick into what thing is beyond wrong into silly. This is about whether we will allow “diversity” issues to be used as wedges to fracture our community, degrade the quality of our work, and draw us away from our duty.

When hackers fail our own standards of meritocracy, as we sometimes do, it’s up to us to fix it from within our own tradition: judge by the work alone, you are what you do, shut up and show us the code. A movement whose favored tools include the rage mob, the dox, and faked incidents of bigotry is not morally competent to judge us or instruct us.

I have been participating in and running open-source projects for a quarter-century. In all that time I never had to know or care whether my fellow contributors were white, black, male, female, straight, gay, or from the planet Mars, only whether their code was good. The SJWs want to make me care; they want to make all of us obsess about this, to the point of having quotas and struggle sessions and what amounts to political officers threatening us if we are insufficiently “diverse”.

Think I’m exaggerating? Read the whole djangoconcardiff thread. What’s there is totalitarianism in miniature: ideology is everything, merit counts for nothing against the suppression of thoughtcrime, and politics is conducted by naked intimidation against any who refuse to conform. Near the end of the conversation djangoconcardiff threatens to denounce Rosario to the board of the Django Software Foundation in the confused, illiterate, vicious idiom of an orc or a stormtrooper.

It has been suggested that djangoconcardiff might be a troll emulating an SJW, and we should thus take him less seriously. The problem with this idea is that no SJW disclaimed him – more generally, that “Social Justice” has reached a sort of Poe’s Law singularity at which the behavior of trolls and true believers becomes indistinguishable even to each other, and has the same emergent effects.

In the future, the hacker whose community standing the SJWs threaten could be you. The SJWs talk ‘diversity’ but like all totalitarians they measure success only by total ideological surrender – repeating their duckspeak, denouncing others for insufficent political correctness, loving Big Brother. Not being a straight white male won’t save you either – Roberto Rosario is an Afro-Hispanic Puerto Rican.

We must cast these would-be totalitarians out – refuse to admit them on any level except by evaluating on pure technical merit whatever code patches they submit. We must refuse to let them judge us, and learn to recognize their thought-stopping jargon and kafkatraps as a clue that there is no point in arguing with them and the only sane course is to disengage. We can’t fix what’s broken about the SJWs; we can, and must, refuse to let them break us.

(Roberto Rosario and Meredith L. Patterson assisted in the composition of this post. However, any errors are the sole responsibility of the author.)

Status signaling and cruelty to betas

I find myself in the embarrassing position of having generated a theoretical insight for a movement I don’t respect very much.

My feelings about the “Red Pill” movement are a lot like my feelings about feminism. Both started out asking important questions about why men and women treat each other badly. Early on, both began to develop some answers that made sense. Later, both movements degenerated – hijacked by whiny, broken, hating people who first edged into outright craziness and then increasingly crossed that line.

But the basic question that motivated the earliest Red-Pill/PUA analysis remains: why do so many women say they want nice guys and then sexually reward arrogant jerks? And the answer has a lot of staying power. Women are instinctive hypergamists who home in on dominance signaling the way men home in on physical pulchritude. And: they’re self-deceivers – their optimal mating strategy is to sincerely promise fidelity to hook a good-provider type while actually being willing to (a) covertly screw any sexy beast who wanders by in order to capture genetic diversity for their offspring, and (b) overtly trade up to a more dominant male when possible.

(This is really complicated compared to the optimal male strategy, which is basically to both find a fertile hottie you think you can keep faithful and screw every other female you can tap without getting killed in hopes of having offspring at the expense of other men.)

What I’ve figured out recently is that there’s another turn of the vise. Sorry, nice-guy betas; you’re even more doomed than the basic theory predicts.

There’a a social-status component to the female game; using it to compete for the attention of fit males. Women are very, very concerned with how their mating value is perceived by others in their social group – they will take extreme measures all the way up to plastic surgery to boost it. Also, female mating value is increased by social status, even though status is not as overwhelmingly important as for males.

Some time back, I tripped over someone else’s realization that this gives women an incentive to be publicly cruel when they reject suitors.

A man courting a woman is implicitly making a status claim: I am good enough for you – in Red Pill terminology, my SMV (sexual market value) meets or exceeds yours. Because other women use male attention to measure SMV and status, such a claim can be threatening to its target because, from a low-status male, it threatens to lower her status, especially if she accepts it.

A woman can deal with this by not merely rejecting a man she evaluates as not being worthy, but publicly insulting him for trying. “How dare you think you’re good enough for me?” is different from a simple “Not interested” because it’s a status defense.

Thus, hot chicks are systematically cruel to beta nerds. It’s a way of socially protecting the proposition that their SMV is high enough to capture a real alpha, and their status among peers.

But – and here’s my insight – it’s even worse than that.

Consider two cases. Bob is slightly lower status than Alice. Ted is much lower status than Alice. Both of them court Alice. She doesn’t think either has SMV to match hers, so her response is to reject both. But: Which one is the bigger status threat?

No, it’s not Ted. The status difference between him and Alice is quite visible to her peers; he can be easily dismissed as just nuts for pitching out of his league. Bob, on the other hand, may look plausible – and the closer to good enough he looks, the more likely it is that the status claim he makes by courting Alice will adjust her status downwards among her peers.

So it’s Bob how will get the cruel, status-defensive rejection, not Ted.

That’s right, guys – being in her league, or nearly so, increases the chance that she’ll have to be nasty to you to protect her game position. The well-spoken, decently groomed nerd is going to get it in the neck from popular hot chick the worst.

However, this analysis does present actionable advice. Because being Bob – being nearly good enough – also increases your odds of being able to raise the SMV she perceives just enough to connect. The advice is: don’t be a social threat. Pitch her privately, not publicly. Give her deniability on your status claim.

At the very least this will give her room to consider whether she likes you without being socially panicked about being seen with the wrong guy.

From kafkatrap to honeytrap

I received a disturbing warning today from a source I trust.

The short version is: if you are any kind of open-source leader or senior figure who is male, do not be alone with any female, ever, at a technical conference. Try to avoid even being alone, ever, because there is a chance that a “women in tech” advocacy group is going to try to collect your scalp.

IRC conversation, portions redacted to protect my informant, follows.

15:17:58 XXXXXXXXXXXX | I'm super careful about honey traps. For a
| while, that's how the Ada Initiative was
| trying to pre-generate outrage and collect
| scalps.
15:18:12 esr | REALLY?
15:18:22 esr | That's perverse.
15:18:42 XXXXXXXXXXXX | Yeah, because the upshot is, I no longer
| can afford to mentor women who are already
| in tech.
15:18:54 esr | Right.
15:19:01 XXXXXXXXXXXX | I can and do mentor ones who are not in
| it, but are interested and able
15:19:21 XXXXXXXXXXXX | but once one is already in... nope
15:20:08 XXXXXXXXXXXX | The MO was to get alone with the target,
| and then immediately after cry "attempted
| sexual assault".
15:23:27 esr | When the backlash comes it's going to be
| vicious. And women who were not part of
| this bullshit will suffer for it.
15:23:41 XXXXXXXXXXXX | I can only hope.
15:25:21 esr | Ah. On the "Pour encourager les autres"
| principle? I hadn't thought of that.
| Still damned unfortunate, though.
15:26:40 XXXXXXXXXXXX | Linus is never alone at any conference.
| This is not because he lets fame go to his
| head and likes having a posse around.
15:26:54 XXXXXXXXXXXX | They have made multiple runs at him.
15:27:29 esr | Implied warning noted.
15:27:34 * | XXXXXXXXXXXX nods

An A&D regular who is not myself was present for this conversation, but I’ll let him choose whether to confirm his presence and the content.

“They have made multiple runs at him.” Just let the implications of that sink in for a bit. If my source is to be believed (and I have found him both well-informed and completely trustworthy in the past) this was not a series of misunderstandings, it was a deliberately planned and persistent campaign to feed Linus to an outrage mob.

Linus hasn’t spoken out about this; I can think of several plausible and good reasons for that. And the Ada Initiative shut down earlier this year. Nevertheless, this report is consistent with reports of SJW dezinformatsiya tactics from elsewhere and I think it would be safest to assume that they are being replicated by other women-in-tech groups.

(Don’t like that, ladies? Tough. You were just fine with collective guilt when the shoe was on the other foot. Enjoy your turn!)

I’m going to take my source’s implied advice. And view “sexual assault” claims fitting this MO with extreme skepticism in the future.