Re: On Smarm

1

It's good. Also, David Denby is the worst. Dave Eggers seems like nearly the worst, and should pretty clearly have not written a novel about... wait, it's about what?


Posted by: Beefo Meaty | Link to this comment | 12- 5-13 1:39 PM
horizontal rule
2

Thanks. I was thinking of how to introduce it via the comments.


Posted by: JP Stormcrow | Link to this comment | 12- 5-13 1:47 PM
horizontal rule
3

Also, rather than tell a story about myself for the fourth time here, I'll just link to the google search. Go team not-earnest!


Posted by: Beefo Meaty | Link to this comment | 12- 5-13 1:48 PM
horizontal rule
4

That's way too long for me to read. This article is all I know about Dave Eggers.


Posted by: Moby Hick | Link to this comment | 12- 5-13 1:54 PM
horizontal rule
5

[Sigh.]


Posted by: Flippanter | Link to this comment | 12- 5-13 1:55 PM
horizontal rule
6

I'm pleased that Eggers wrote a novel about something without bothering to inform himself about the dreary day-to-day facts of the matter; it shows a healthy disregard for facts suitable for a writer of fiction. What would we prefer, that Eggers fall into careless habits of accuracy?


Posted by: nosflow | Link to this comment | 12- 5-13 1:56 PM
horizontal rule
7

Between the Eggers rant and Conan's speech, Harvard's class of 2000 sure got some inexplicably still-shareable Gen X wisdom.


Posted by: Criminally Bulgur | Link to this comment | 12- 5-13 1:57 PM
horizontal rule
8

Also, I was totally going to post a link to this article, but I was going to say more about it, too, in the body of the post. This just shows how much greater a master of the blog form ogged is than I am.


Posted by: nosflow | Link to this comment | 12- 5-13 1:57 PM
horizontal rule
9

In another realm of discourse, had just read this smarmy response to Elizabeth Warren's response to the recent smarmy Third Way WSJ OpEd and thought the essay fit it well.


Posted by: JP Stormcrow | Link to this comment | 12- 5-13 2:00 PM
horizontal rule
10

Is it possible to hate both Dave Eggers and Gawker Media deeply?


Posted by: Robert Halford | Link to this comment | 12- 5-13 2:01 PM
horizontal rule
11

I'm here to tell you that yes, yes it is.


Posted by: Robert Halford | Link to this comment | 12- 5-13 2:01 PM
horizontal rule
12

Or, put differently, most "snark" is just another version of what is called here "smarm." It's all self-promoting bullshit all the way down.


Posted by: Robert Halford | Link to this comment | 12- 5-13 2:02 PM
horizontal rule
13

12 confuses me.


Posted by: nosflow | Link to this comment | 12- 5-13 2:03 PM
horizontal rule
14

I thought the critique was that the smarmy, pro-earnestness folks were basically just creating a "personal brand" and self-promoting. Which, fair enough. But an enormous amount of the Gawker-media type stuff is just the same thing with a different style, instead of fake earnestness we have fake controversy and feigned outrage designed to drive page views and self-promote. Basically my view is that everyone, possibly including me, should just shut the fuck up.


Posted by: Robert Halford | Link to this comment | 12- 5-13 2:08 PM
horizontal rule
15

I'm confused about the word choice. Like the bit about Maria Bartiromo acting shocked that someone could call JP Morgan Chase "corrupt" -- is that "smarm"? It's a thing that is very common, an emphasis on civility over correctness, but I'm not sure "smarm" is the right word.


Posted by: essear | Link to this comment | 12- 5-13 2:10 PM
horizontal rule
16

I think you missed part of the critique of smarm, that it's often a way to block substantive criticism by defining it as unacceptably rude or angry or harsh.


Posted by: LizardBreath | Link to this comment | 12- 5-13 2:10 PM
horizontal rule
17

To refrain mutually from injury, from violence, from exploitation, and put one's will on a par with that of others: this may result in a certain rough sense in good conduct among individuals when the necessary conditions are given (namely, the actual similarity of the individuals in amount of force and degree of worth, and their co-relation within one organization). As soon, however, as one wished to take this principle more generally, and if possible even as the fundamental principle of society, it would immediately disclose what it really is--namely, a will to the denial of life, a principle of dissolution and decay.


Posted by: OPINIONATED FRIEDRICH NIETZSCHE | Link to this comment | 12- 5-13 2:13 PM
horizontal rule
18

16 - Oh, that part I agree with. But there's also a countervailing tendency to overvalue the snarky rant. Neither one is really the same thing as substantive criticism -- it's all about self promotion, not actually engaging with the topic.


Posted by: Robert Halford | Link to this comment | 12- 5-13 2:14 PM
horizontal rule
19

This thing, which may or may not be "smarm", is also super-common in academia and is the main reason why I now find most conferences completely unbearable.


Posted by: essear | Link to this comment | 12- 5-13 2:15 PM
horizontal rule
20

What about smarky? Smarmy and snarky all together in one great package. Dude, why you gotta be so smarky?

https://twitter.com/morningmoneyben/status/408652481384955905


Posted by: peep | Link to this comment | 12- 5-13 2:16 PM
horizontal rule
21

But there's also a countervailing tendency to overvalue the snarky rant.

But, but, but ... Mencken! Jimmy Breslin! Mencken!


Posted by: Flippanter | Link to this comment | 12- 5-13 2:17 PM
horizontal rule
22

"Smark" is taken.


Posted by: Flippanter | Link to this comment | 12- 5-13 2:18 PM
horizontal rule
23

How did you know the comment I was about to post, Flippanter?


Posted by: Beefo Meaty | Link to this comment | 12- 5-13 2:21 PM
horizontal rule
24

15: I hadn't ever seen 'smarm' used for this kind of thing, but I also didn't have a neat word for it, so smarm works fine for me. "Dodging substantive engagement by hiding behind a false expectation of civility that puts significant and important subject matter areas out of bounds as too uncivil to discuss" is the sort of thing I'd come up with, but coining punchy terms has never been my strong point.


Posted by: LizardBreath | Link to this comment | 12- 5-13 2:22 PM
horizontal rule
25

It doesn't matter how I know, jabroni!


Posted by: Flippanter | Link to this comment | 12- 5-13 2:22 PM
horizontal rule
26

I'm confused about the word choice.

Yeah, only some of these are smarmy. Also, pace Halford, I didn't think the overall critique was that smarm = personal brand creation (though I also think that Halford is conflating the thing described in the odious fast company piece with "having a voice" or being distinctive in any way), but rather that smarminess, or whatever you call it, is a particular distinctive way of acting which is dishonest and bad.


Posted by: nosflow | Link to this comment | 12- 5-13 2:23 PM
horizontal rule
27

I feel like "bullshit sanctimoniousness" flows pretty well.


Posted by: Beefo Meaty | Link to this comment | 12- 5-13 2:24 PM
horizontal rule
28

I just want to reserve my right to also hate Gawker Media. Manufacturing outrage for page views is also deeply bullshit!!


Posted by: Robert Halford | Link to this comment | 12- 5-13 2:25 PM
horizontal rule
29

24: But "smarm" already means something, and it isn't that. Or at least I didn't think it was. "Smarm" implies unctuousness and mostly, to me, suggests people trying to curry favor with other more powerful people, whereas this other thing is just as often a tool of powerful people to shut down discussion.


Posted by: essear | Link to this comment | 12- 5-13 2:25 PM
horizontal rule
30

The whole thing in the Believer about negative reviews and the condemnation of snark was, IIRC, a large part of the brief against said rag by the successors of n. I feel as though it's a large part of what passes for literary culture in this once-proud city; witness the parade of crappiness Litquake, where everyone is so proud of themselves for having taken part, and nevermind the quality of what they took part in.


Posted by: nosflow | Link to this comment | 12- 5-13 2:26 PM
horizontal rule
31

You're allowed to hate Gawker, Rob.


Posted by: nosflow | Link to this comment | 12- 5-13 2:26 PM
horizontal rule
32

But also, is this whatever-it-is we're talking about particularly new? I mean hasn't this always just been a pretty standard tool in the arsenal of anybody who wants to deflect criticism.


Posted by: Robert Halford | Link to this comment | 12- 5-13 2:27 PM
horizontal rule
33

27: It does, but isn't it redundant? Whatever happened to Jedediah Purdy, anyway?


Posted by: Jesus McQueen | Link to this comment | 12- 5-13 2:29 PM
horizontal rule
34

I had nothing to do with Litquake.


Posted by: Bave | Link to this comment | 12- 5-13 2:29 PM
horizontal rule
35

30 -- OH, now I get it. I agree! That thing, whatever it is, is just so deeply ingrained in current literary culture, but I must say my prejudice is that it's a particularly strong and off-putting element of many aspects of your fair city.

33 -- he's a law prof (and a friend or friendquaintance of a bunch of people here, I believe).


Posted by: Robert Halford | Link to this comment | 12- 5-13 2:31 PM
horizontal rule
36

I've always thought smarm meant sort of a disingenuous friendliness. A property associated with used car salesmen and televangelists. Defining it the way the article does* seems to me just wrong.

*In all fairness I haven't read the article, just the eighty gazillion blog posts approvingly linking to it.


Posted by: togolosh | Link to this comment | 12- 5-13 2:34 PM
horizontal rule
37

he's a law prof

Irony ftw.


Posted by: Jesus McQueen | Link to this comment | 12- 5-13 2:34 PM
horizontal rule
38

Halford didn't read the whole thing.


Posted by: ogged | Link to this comment | 12- 5-13 2:40 PM
horizontal rule
39

33 27: It does, but isn't it redundant?

You could shorten it to "sanctimon".


Posted by: essear | Link to this comment | 12- 5-13 2:41 PM
horizontal rule
40

38: Any yet opined. Shocking.


Posted by: JP Stormcrow | Link to this comment | 12- 5-13 2:44 PM
horizontal rule
41

20: Ha! It was Ben White's response to EW's response to TW I was pointing to in 9.


Posted by: JP Stormcrow | Link to this comment | 12- 5-13 2:45 PM
horizontal rule
42

Could anyone possibly read all of that? Not possible.


Posted by: Robert Halford | Link to this comment | 12- 5-13 2:49 PM
horizontal rule
43

It was really long and repetitive -- I was agreeing with it in the beginning, and then just scanning to try to be sure it didn't go someplace weird that I wouldn't want to agree with.


Posted by: LizardBreath | Link to this comment | 12- 5-13 2:50 PM
horizontal rule
44

Nelson Mandela just died of outrage over your failure to finish the article before commenting. Way to go.


Posted by: ogged | Link to this comment | 12- 5-13 2:52 PM
horizontal rule
45

I actually didn't really buy where he went with it towards the end, but that's fine. Look, Thumper!


Posted by: Beefo Meaty | Link to this comment | 12- 5-13 2:53 PM
horizontal rule
46

The author wants to publish it as a large-print little book, set out near the cash registers at Barnes & Noble. You need a certain word count to do that.


Posted by: Bave | Link to this comment | 12- 5-13 2:53 PM
horizontal rule
47

Tuesdays with Smarmberries


Posted by: Beefo Meaty | Link to this comment | 12- 5-13 2:55 PM
horizontal rule
48

Hey, ogged wasn't kidding. Halford actually killed Mandela! That's messed up.


Posted by: Beefo Meaty | Link to this comment | 12- 5-13 2:58 PM
horizontal rule
49

46: Well, sure for the book, but you don't need to pad an article that much.


Posted by: peep | Link to this comment | 12- 5-13 2:58 PM
horizontal rule
50

That's probably only the fourth or fifth worst thing Halford has ever done.


Posted by: ogged | Link to this comment | 12- 5-13 3:01 PM
horizontal rule
51

What can I say, the guy ended apartheid but he shouldn't have downloaded Iron Man III.


Posted by: Robert Halford | Link to this comment | 12- 5-13 3:07 PM
horizontal rule
52

51: First you kill him then you slander him???


Posted by: peep | Link to this comment | 12- 5-13 3:11 PM
horizontal rule
53

Speaking of racist dickheads who were dicks to Mandela, from that great Perlstein piece on MIA/POWs:

But how could there be proof that men shot down over jungles or the Gulf of Tonkin or the South China Sea were "really" dead? And so the "issue" endured. Governor Ronald Reagan, in Singapore as a special presidential representative for a trade deal, said that if North Vietnam didn't return the POWs and MIAs supposedly still being held, "bombing should be resumed." He accused liberals in Congress seeking to ban further military action in Southeast Asia of taking away "the power to sway those monkeys over there to straighten up and follow through on the deal."


Posted by: JP Stormcrow | Link to this comment | 12- 5-13 3:13 PM
horizontal rule
54

This thing, which may or may not be "smarm", is also super-common in academia and is the main reason why I now find most conferences completely unbearable.

Really? People accuse each other of being uncivil when the uncivil person is actually pointing out a real problem?


Posted by: heebie-geebie | Link to this comment | 12- 5-13 3:16 PM
horizontal rule
55

53.1 not referring to Halford in case there was any confusion.


Posted by: JP Stormcrow | Link to this comment | 12- 5-13 3:16 PM
horizontal rule
56

Oh come on white people, don't you think changing your profile picture to a picture of Mandela is a little bit...something?


Posted by: Mister Smearcase | Link to this comment | 12- 5-13 3:20 PM
horizontal rule
57

56: Smarmy?


Posted by: peep | Link to this comment | 12- 5-13 3:24 PM
horizontal rule
58

Ugggh, I have only read about 2/3 of the essay but I find it grating. I'm not entirely sure why, but my first reaction is that it piles on example and example (to the point of tedium) without serving to refine the concept very effectively, because it never asks what the limits of the definition are, or looks for an edge case.

I get it, it's easy to find lots of examples of people cloaking self-interest (or service to the interest of others) in sanctimony or generally tut-tutting their critics. What it doesn't ask is, "is this anything new?", "is there any reason to think that 'smarm' is more common now than it used to be?", or "does 'smarm' actually work, or are most of the examples pretty transparent dissembling?"

Furthermore, I don't think "smarm," by that definition, is the main target of snark. I think of snark as targeting actual conventions of journalistic style, like false even-handedness ("shape of the earth: opinions differ") or ruling certain subjects outside of the domain of polite or adult conversation (in that sense you might think of Dan Savage as the perfect example of how snark opened up space for earnest discussion of topics which had been marginalized) rather than appeals to the idea of conventionality.


Posted by: NickS | Link to this comment | 12- 5-13 3:34 PM
horizontal rule
59

People accuse each other of being uncivil when the uncivil person is actually pointing out a real problem?

More often people just avoid pointing out problems at all because they don't want to be perceived as being confrontational.

But yes, it definitely happens. I could give examples but they wouldn't make a lot of sense unless I wrote very long comments with lots of context.

Actually, you could Google up examples, although not very interesting ones: e.g. there are all kinds of places on the web where someone explains that loop quantum gravity is a completely wrong, failed research field, and then some of its proponents show up and throw a fit about how rude the critics are. The more interesting examples are things that are more mainstream research, are wrong, and keep getting worked on and funded and cited because people aren't willing to say "bullshit" loudly enough.


Posted by: essear | Link to this comment | 12- 5-13 3:35 PM
horizontal rule
60

Huh. Granted, I only know a very teeny corner of the math community, but that is so very different than how they operate.


Posted by: heebie-geebie | Link to this comment | 12- 5-13 3:37 PM
horizontal rule
61

I would argue that smarm and snark are both forms of bullshit -- they are both reactions which are essentially non-substantive. To call a substantive but negative response snark is itself to commit snark.


Posted by: text | Link to this comment | 12- 5-13 4:01 PM
horizontal rule
62

35.2 - yes, he's a lovely human! He's editing a volume on neoliberalism and the law right now, the draft introduction of which may be available if anyone's interested. He's been active in NC around the recent Republican revolt, and he posts a lot of pictures to FB of giant mushrooms he finds in the woods.

He also has a good sense of humor about For Common Things. (It's worth noting that FCT opposed irony as a default viewpoint, but celebrated it in the service of justice, a la Swift.) The article isn't fair to him as a person -- especially the bit about winding up a law prof -- but it's certainly fair on the way his book was roped into the culture at large.


Posted by: k-sky | Link to this comment | 12- 5-13 4:20 PM
horizontal rule
63

I'd say that I was glad to see the loathsome and inexplicably not banished to die on a nice floe alone and in pain Lee Siegel show up in this, but that would be content-free snark.


Posted by: snark out | Link to this comment | 12- 5-13 4:20 PM
horizontal rule
64

Fucking iPhones.


Posted by: Snarkout | Link to this comment | 12- 5-13 4:21 PM
horizontal rule
65

The whole thing in the Believer about negative reviews and the condemnation of snark was, IIRC, a large part of the brief against said rag by the successors of n

A Regressive Avant-Garde. Better done than the Scocca, IMHO.


Posted by: k-sky | Link to this comment | 12- 5-13 4:21 PM
horizontal rule
66

This thread reminded me to respond to the guy who sent me a complaint the other day about how I never refer to his work in my talks. I tried to be scrupulously polite but also tell him why I think everything he does is nonsense. Of course, I also forwarded his email to some friends with snarky comments appended.


Posted by: essear | Link to this comment | 12- 5-13 4:27 PM
horizontal rule
67

If we're going to banish people to die on a floe, it might as well be a nice floe.


Posted by: essear | Link to this comment | 12- 5-13 4:28 PM
horizontal rule
68

As to the argument itself, I think it's aiming in the right direction but symptomatic of a general disconnect in the culture between creators and critics. Yes, the impulse towards generosity and enthusiasm gets subsumed into an anti-critical posture that stifles debate, but it's also possible to examine the strengths of a thing, and encouragement is not ipso facto coddling. It's interesting to me that Eggers's effect on literary culture may be one of charitability to a fault, but that he's gone so far as to actually create a charitable literary culture, with good works like 826 that create literacy programs and literary works like Zeitoun that support activism around Katrina.

As the friend who introduced me to Jed pointed out in an email discussion, the argument he's having is going on much more productively in n+1 and the New Inquiry and Jacobin. Still, I'm mostly sympathetic.


Posted by: k-sky | Link to this comment | 12- 5-13 4:29 PM
horizontal rule
69

If this (last line) is snark, I want to see more of it.


Posted by: essear | Link to this comment | 12- 5-13 4:42 PM
horizontal rule
70

Smarm is when you suck up to people in power. That isn't what anybody in that article is doing really. People like Eggers are just being positive and trying new things and new ventures. Others are just prissily complaining about criticism directed to themselves.

It probably isn't a good idea to be negative about things. At least culturally, bad things are more avoidable than ever.

There was a certain sense in making fun of a Bob Hope special or something back when it might actually been the most entertaining thing on TV. There were only three channels, but there are a lot more options now. Just look for something that you do like.


Posted by: lemmy caution | Link to this comment | 12- 5-13 7:31 PM
horizontal rule
71

It's lemmy! Lemmy is the enemy! GET HIM!


Posted by: Beefo Meaty | Link to this comment | 12- 5-13 7:34 PM
horizontal rule
72

That article is totally pointless and I wish it didn't exist.


Posted by: dz | Link to this comment | 12- 5-13 7:49 PM
horizontal rule
73

I don't think I need to read the whole thing to say NickS is exactly right in 58.


Posted by: fake accent | Link to this comment | 12- 5-13 8:11 PM
horizontal rule
74

I love how Halford didn't get this until neb framed it as a criticism of San Francisco.


Posted by: teofilo | Link to this comment | 12- 5-13 8:17 PM
horizontal rule
75

73 gets it exactly right.


Posted by: parsimon | Link to this comment | 12- 5-13 8:22 PM
horizontal rule
76

It's a good piece, but does suffer from the problems NickS notes (and from undue focus on Eggers--that ends up being a real distraction). At some point, he needs a good, condensed definition of the thing--you gotta hen before you pros. But it's not like it's impossible to see what he's getting at: a culture of unctuous positivity devalues truth and insulates people from valid criticism. All his examples more or less fit, but it's not a tight piece.


Posted by: ogged | Link to this comment | 12- 5-13 8:34 PM
horizontal rule
77

I do think, though, that he identified a bunch of different things -- the I-used-to-punch-up-but-now-that-I'm-on-top-I-decry-the-new-epidemic-of-punching-up is the most interesting, but the pearl-clutchery is also notable -- that don't really go together but that writers he doesn't like do, and he sort of dumps them together. I thought Sedaris in particular doesn't fit. Dislike the guy for making shit up and/or This American Lifeism, sure, but he can be catty as hell.


Posted by: Beefo Meaty | Link to this comment | 12- 5-13 8:37 PM
horizontal rule
78

The basic thing he identifies seems clear enough, whatever words you use to describe it, but it doesn't seem particularly more characteristic of the era in which "snark" became a word in common use than before. Go back far enough and sure, there's probably a point where it becomes identifiably new. I'd rather read about that than see someone dredge up Lee Siegel's writing just to rip into it.


Posted by: fake accent | Link to this comment | 12- 5-13 8:44 PM
horizontal rule
79

To be clear, I'm not saying there's ever a good reason dredge up Lee Siegel's writing. No one should be asked to read any of it for any reason.


Posted by: fake accent | Link to this comment | 12- 5-13 8:46 PM
horizontal rule
80

Yeah, that guy sucks balls. Can't even front.


Posted by: Sprezzatura | Link to this comment | 12- 5-13 8:48 PM
horizontal rule
81

I dunno, I thought there were some interesting seeds of thoughts (particularly in the last 1/3 of the essay) but it wasn't the piece I was hoping for when I clicked.

To me it's inarguable that our society both undermines genuine, earnest praise (by publishing so much empty praise) and uses civility as a cudgel (mostly against less-powerful people, to punish them for bringing up uncomfortable topics).

That said, the question is what to do about it. I deal with it on a personal level by editing out garbage praise when I can, and pushing back against civility bullying in a way that explicitly calls out what the person is doing. That way, even if the group goes along with it, there is a voice naming it for what it is.

I'm at the point in my life where it feels a lot more unforgiveable to tolerate rhetorical jujitsu than it does to be accused of being unpleasant.


Posted by: Witt | Link to this comment | 12- 5-13 8:49 PM
horizontal rule
82

One more thought. Snark can be helpful when it cuts to the chase. Some people can hear a succinct, snarky critique when a more reasoned one just sounds like a wall of words to them.


Posted by: Witt | Link to this comment | 12- 5-13 9:04 PM
horizontal rule
83

It's petty, but I'm in favor of any writing that gives Jedediah Purdy a hard time, because I viscerally recall reading his original op-ed piece and being deeply aggravated by it.


Posted by: Nathan Williams | Link to this comment | 12- 5-13 9:05 PM
horizontal rule
84

Funny. I liked the essay, but Witt did an awesome job of distilling the point that I most identified with, which highlighted the reason I thought it was good but not awesome. Rhetorical jujitsu is really the thing that should be argued against, especially the meta version, i.e. Your criticism isn't valid because criticism itself is rude.


Posted by: F | Link to this comment | 12- 5-13 9:13 PM
horizontal rule
85

particularly in the last 1/3 of the essay

<whine>Nooooo. I don't want to read that far.</whine>


Posted by: fake accent | Link to this comment | 12- 5-13 9:16 PM
horizontal rule
86

Some people can hear a succinct, snarky critique when a more reasoned one just sounds like a wall of words to them.

So what you're saying is that we shouldn't anthropomorphize people?


Posted by: Sidney Morgenbesser | Link to this comment | 12- 5-13 9:32 PM
horizontal rule
87

I like the piece. Concern about "smarm" is less annoying than concern about "tone policing" that I see all the time when some blogger whose schtick is to be angry all the time gets angry when someone gets angry at her for being angry all the time. Guess what, nobody likes hearing a stranger vent. Unless the stranger shares our experiences and is therefore (literally) speaking for us.


Posted by: Cryptic ned | Link to this comment | 12- 5-13 11:15 PM
horizontal rule
88

If you don't like angry feminist bloggers, Ned, maybe you should just stop reading them.


Posted by: teofilo | Link to this comment | 12- 5-13 11:59 PM
horizontal rule
89

I'm fine with strangers venting if they're awesome at it like Lewis Black.


Posted by: gswift | Link to this comment | 12- 6-13 12:17 AM
horizontal rule
90

56 our Prime Minister, who couldn't remember whether or not he was in favour of the '81 Springbok Tour (people fighting in the streets and stadiums, planes dropping flour bombs on test matches, rioting in all major cities etc) is now dropping every fucking platitude about Mandela and it's so fucking gross.


Posted by: Keir | Link to this comment | 12- 6-13 2:42 AM
horizontal rule
91

This isn't an angry feminist blogger, it's an angry science blogger.


Posted by: Cryptic ned | Link to this comment | 12- 6-13 8:29 AM
horizontal rule
92

Maybe they got confused about the phrase "mad scientist".


Posted by: essear | Link to this comment | 12- 6-13 8:30 AM
horizontal rule
93

92: If he had the courage to say "I supported it but I was wrong and only 20 years old", (with him being a politically active Muldoon supporter, seems to be correct) it wouldn't seem so pathetic.

On our side of the Pacific, we had our most sainted president calling the Anti-Apatheid Act "immoral" and "utterly repugnant." Funny that that isn't part of the remembrance.


Posted by: dalriata | Link to this comment | 12- 6-13 8:45 AM
horizontal rule
94

Oh come on white people, don't you think changing your profile picture to a picture of Mandela is a little bit...something?

Uh oh. I'm wearing an ANC shirt.


Posted by: | Link to this comment | 12- 6-13 8:53 AM
horizontal rule
95

96 was me.


Posted by: Sir "I ended apartheid" Kraab | Link to this comment | 12- 6-13 8:54 AM
horizontal rule
96

David Cameron is the fucking visual definition of smarmy, the platonic ideal. The crimson tide outbursts just highlight it through counterpoint.


Posted by: Alex | Link to this comment | 12- 6-13 9:04 AM
horizontal rule
97

Oh, thanks for the Greg Wilson link, Sifu. He's playing the Nest and XOYO in quick succession and I can use that as propaganda/briefing material to get my girlfriend to go.


Posted by: Alex | Link to this comment | 12- 6-13 9:07 AM
horizontal rule
98

Republicans fighting Obamacare are like Nelson Mandela fighting apartheid according to Rick Santorum.


Posted by: Moby Hick | Link to this comment | 12- 6-13 9:08 AM
horizontal rule
99

I would not object to Rick Santorum being locked up on an island for 28 years.


Posted by: togolosh | Link to this comment | 12- 6-13 9:25 AM
horizontal rule
100

I might object to the Specials doing a pro-Santorum song.


Posted by: Natilo Paennim | Link to this comment | 12- 6-13 9:44 AM
horizontal rule
101

And, in the movie version, you definitely can't have Idris Elba playing Santorum. Bill Murray maybe?


Posted by: Stanley | Link to this comment | 12- 6-13 9:57 AM
horizontal rule
102

Wait, I know: Nic Cage.


Posted by: Stanley | Link to this comment | 12- 6-13 10:00 AM
horizontal rule
103

104: Rick Moranis.

So, in 2051, after 28 years in Guantanamo, Santorum is freed, and negotiates an end to Obamacare, and becomes President of the once-again free to be uninsured USA.

No, thank you.


Posted by: peep | Link to this comment | 12- 6-13 10:17 AM
horizontal rule
104

I don't know about the rest, but 105.1 is obviously right.


Posted by: Moby Hick | Link to this comment | 12- 6-13 10:19 AM
horizontal rule
105

Here is an article about a bunch of positive people.

http://www.newyorker.com/reporting/2013/10/14/131014fa_fact_heller

I am mainly just jealous.


Posted by: lemmy caution | Link to this comment | 12- 6-13 10:36 AM
horizontal rule
106

Rick Santorum will be played by Alan Ruck, obviously, and it looks like he could use the work.


Posted by: Jackmormon | Link to this comment | 12- 6-13 11:18 AM
horizontal rule
107

102 - Ain't he cute? / No he ain't / Just another burden / On the welfare state


Posted by: snarkout | Link to this comment | 12- 6-13 11:40 AM
horizontal rule
108

One of the commenters on Tim Burke's Mandela post seems intent on demonstrating smarm.


Posted by: fake accent | Link to this comment | 12- 6-13 12:40 PM
horizontal rule
109

107: Man, does anyone know who is responsible for the endless New Yorker love letters to Silicon Valley? ("Sometimes I love you and sometimes I worry about you, but I never stop thinking about you, baby.") Are they getting anything back?


Posted by: lurid keyaki | Link to this comment | 12- 6-13 1:19 PM
horizontal rule
110

@110: I saw that as well. Of course there is absolutely nothing more uncivil than noting that the folks who wanted Mandela to stay locked up did, in fact, want him to stay locked up.

And naturally when someone whose entire life has been defined by his role in a political struggle dies, the one thing we mustn't do is politicize his death.


Posted by: AcademicLurker | Link to this comment | 12- 6-13 1:26 PM
horizontal rule
111

I couldn't make it past the first page of that circle jerk in 107.


Posted by: gswift | Link to this comment | 12- 6-13 1:32 PM
horizontal rule
112

The main thing that struck me about the article in 107 is that I live so close yet so far from those people. My sister lives 70 miles closer, and still so far.


Posted by: Megan | Link to this comment | 12- 6-13 1:53 PM
horizontal rule
113

114: Judging distance to an alternate reality is always going to be tough.


Posted by: gswift | Link to this comment | 12- 6-13 2:07 PM
horizontal rule
114

An acquaintance wrote this at The Other Place and I thought it was pretty much spot-on, except for his judgment about which error is more common:

Nelson Mandela will be remembered for many things; primarily, I suspect, misremembered as someone far less radical than he was. But among the things he should be remembered as is someone who embodied the Weberian ethic of responsibility. Too many people of good intentions reject violence prejudicially; a more obviously disastrous, though for that reason less common, error is to romanticize violence or to treat it as spiritually ennobling. Mandela was the opposite of a cynic, but he had the cynic's one virtue, in that he understood that both coercion and compromise are tactics, not goals in themselves.

Posted by: Benquo | Link to this comment | 12- 6-13 2:21 PM
horizontal rule
115

116: At least he didn't mention The Sound of Music Live!.


Posted by: Stanley | Link to this comment | 12- 6-13 2:30 PM
horizontal rule
116

This certainly isn't on the level of Obamacare = death = apartheid, but during the divestment campaign at DFHC a handful of Republicans responded to the shanties we built (it was a thing at the time, youngsters) by building an O!Berlin Wall (actual punctuation used, not google-proofing) because, see, our cherry-picking of symbolic structures demonstrated that leftists don't object to oppression by other leftists.


Posted by: Sir Kraab | Link to this comment | 12- 6-13 2:31 PM
horizontal rule
117

I quite like 116.


Posted by: Sir Kraab | Link to this comment | 12- 6-13 2:32 PM
horizontal rule
118

It took quite a while to figure out 118 because I assumed the "O" was added to "Berlin Wall" to make fun of Obama.


Posted by: Crypticneed | Link to this comment | 12- 6-13 2:41 PM
horizontal rule
119

At my college some dufus tried to burn down the shanties late at night apparently not realizing that people stayed in the shanties overnight.


Posted by: lemmycaution | Link to this comment | 12- 6-13 3:03 PM
horizontal rule
120

That's an optimistic view of human nature.


Posted by: Moby Hick | Link to this comment | 12- 6-13 3:18 PM
horizontal rule
121

I think Ned just improved his moniker with a typo.


Posted by: gswift | Link to this comment | 12- 6-13 3:56 PM
horizontal rule
122

107.last gets it basically right for me, too. I found this amusing:

I didn't understand, for instance, why suddenly every business wanted to be a "startup." I didn't understand how people like Hwin appeared to float above the exigencies of career.

Actually, no, I think you got it, a few paragraphs back:

He went to Stanford on a full financial-aid package and majored in psychology. That was where he became interested in business, and also in the Bay Area tech scene. He built one of the first "viral, spammy" Facebook apps, called Quizzes, and it gained fifteen million users. He sold it, but he wasn't proud of it, and so he used the money to found another company, Damntheradio--a Facebook-marketing platform ultimately used by Lady Gaga and other musicians looking to harness a social-media fan base. He sold that, in 2010, for a combination of stock and cash reported to be in the low seven figures.

When you have wealth in the seven figures as a 25-year-old, you don't just appear to "float above the exigencies of career", you actually are above them. That's what wealth means, and a large part of why people want it. (All the more so when this comes not from, say, a lottery jackpot but from skills that promise more where that came from if need be.)

(Why yes, I have not yet found a job, how did you guess?)


Posted by: x.trapnel | Link to this comment | 12- 6-13 4:21 PM
horizontal rule
123

And naturally when someone whose entire life has been defined by his role in a political struggle dies, the one thing we mustn't do is politicize his death.

Moreover, to erase all of the details of that political struggle (including, for example, the reasons why that struggle had been necessary in the first place) in order to depict the deceased as an emblem of reconciliation, as a hero for having risen 'above politics' ... is not at all to politicize his death.


Posted by: Just Plain Jane | Link to this comment | 12- 6-13 9:13 PM
horizontal rule
124

(Why yes, I have not yet found a job, how did you guess?)

Get that police app in so you can do things like get bit by a guy after you tell him to stop riding his bike on the sidewalk and then have some of the other stellar citizens in the area turn it into a Jerry Springeresque donnybrook.


Posted by: gswift | Link to this comment | 12- 7-13 11:07 AM
horizontal rule
125

124, 126: Seriously, x, do it do it do it, and not just because I think it would be funny. You're kind and reasonable, which is what we want in cops. Academia didn't suit you, in a way that makes me think that a more in-the-moment and less months-to-years-long self-motivated deadlines job might easily work better for you. And you'd have good stories to tell.


Posted by: LizardBreath | Link to this comment | 12- 7-13 11:16 AM
horizontal rule
126

124, 126, 127

Seconded (or thirded, whatever)
This in particular:
Academia didn't suit you, in a way that makes me think that a more in-the-moment and less months-to-years-long self-motivated deadlines job might easily work better for you.
brings out the "knowing now my strengths and weakness and if I could do it all over again' feeling in me.


Posted by: Barry Freed | Link to this comment | 12- 7-13 11:26 AM
horizontal rule
127

I disagree. The things that make a person like or love academia can easily remain likes or loves (they have for me!), and pretending that you're not, or never were, or are no longer, deeply attracted to those things can be a mistake. They can be found outside the academy; no need to enter into something that's the polar opposite.


Posted by: parsimon | Link to this comment | 12- 7-13 11:47 AM
horizontal rule
128

a more in-the-moment and less months-to-years-long self-motivated deadlines job might easily work better for you

Patrol is glorious that way. You take your calls, look for people to arrest, write your reports, and walk away at the end with nothing hanging over your head and each day you start with a clean slate. The detective gig is only slightly less that way and still pretty awesome in that regard.


Posted by: gswift | Link to this comment | 12- 7-13 12:00 PM
horizontal rule
129

Being a detective has always seemed like an awesome job, but there's the whole now you're The Man thing. And you have to do regular police duty first, right?

FBI seems like a good gig, too, but then you really are the man, and I think they have an age cutoff for new agents. (No idea how old trapnel is, or who he is, or what he does, or even, really, if he's a serial killer.)


Posted by: ogged | Link to this comment | 12- 7-13 12:17 PM
horizontal rule
130

This is roughly how I conceive of x.trapnel on the job.


Posted by: LizardBreath | Link to this comment | 12- 7-13 12:19 PM
horizontal rule
131

no need to enter into something that's the polar opposite.

Yes, often that's exactly what's needed. That feeling Barry mentions I still get in the sense of why on earth did it take me so godamn long to figure these things out about myself. That's exactly why I'm egging Trapnel on to do it. I was 32 when I started back in '08 and if I hadn't there's a chance I'd still be futzing around in some dead end corporate gig I didn't really enjoy and ready to blow my brains out feeling like I'd wasted my life and potential. The job is interesting and satisfying and I have financial stability and loads of free time to pursue my other interests. This is the happiest I've ever been in my adult life and I'm pretty sure it'd be a huge improvement for him too.


Posted by: gswift | Link to this comment | 12- 7-13 12:23 PM
horizontal rule
132

Oh, I understand, gswift, certainly. It's up to trapnel whether he thinks that kind of thing would get his groove on. The things I've seen him get really passionate about -- just here, obviously, I don't know about elsewhere -- are, remain, intellectual in nature. trapnel needs a job like ttaM's, I ... speculate.

(Hey, x.trapnel: we're talking about you in front of your back!)

FBI would be awesome if you could be, like, Fox Mulder.


Posted by: parsimon | Link to this comment | 12- 7-13 12:31 PM
horizontal rule
133

||
Speaking of smarm, FFS Yggles should be ashamed.
|>


Posted by: Josh | Link to this comment | 12- 7-13 12:32 PM
horizontal rule
134

Yes, everyone has to work the street first but I thought patrol was great and am glad I didn't rush to get out of it. It did take a bit wrapping my head around being The Man but there's loads of discretion as to how you do the job. And in a town like SF I'm pretty sure no one's going to be trying to push things like weed arrests. A lot of the federal stuff is overrated. Federal gigs tend to move you around and any idea that the Feds are all high speed cool guys and shit disappears the minute you actually work with them on something. You're way more likely to be doing fun enjoyable work in a metro PD.


Posted by: gswift | Link to this comment | 12- 7-13 12:33 PM
horizontal rule
135

why on earth did it take me so godamn long to figure these things out about myself

Still, this is a really compelling train of thought. Maybe I should have been a goddamn fire-spinner, or eagle trainer, or mountain guide. It's absolutely worth entertaining these questions: right livelihood.


Posted by: parsimon | Link to this comment | 12- 7-13 12:36 PM
horizontal rule
136

Academia didn't suit you, in a way that makes me think that a more in-the-moment and less months-to-years-long self-motivated deadlines job might easily work better for you.

This is very true, but I'm hoping I can get that from Agile development practices, with less internal dissonance of being The Man, and, most importantly, less conflict. I'm incredibly conflict-averse.

TAing an 'HTML for beginners' workshop right now so can't really respond, but please do feel free to continue talking about me and planning out my life.


Posted by: x.trapnel | Link to this comment | 12- 7-13 12:43 PM
horizontal rule
137

135: He should crawl under a fucking rock for the next five years, until he stops being 12 years old.


Posted by: parsimon | Link to this comment | 12- 7-13 12:43 PM
horizontal rule
138

I'm incredibly conflict-averse.

Ah, that would be a problem in patrol because as you can imagine, conflict is pretty much why people call the cops.


Posted by: gswift | Link to this comment | 12- 7-13 12:45 PM
horizontal rule
139

"But why would I buy some socks or a no-stick frying pan or a coffee maker at Walmart when Amazon Prime would ship almost anything to my door in 36 hours?"


Posted by: Cryptic ned | Link to this comment | 12- 7-13 12:48 PM
horizontal rule
140

TrapnelCop would be great. Plus SF has such a great cop tradition: Dirty Harry, Karl Malden, Bullitt. I wonder if they train you to race up and down hills in a 68 Mustang.


Posted by: Robert Halford | Link to this comment | 12- 7-13 12:48 PM
horizontal rule
141

I think I'd be the only cop around here who would rather be patrolling the terrifying cities of 10,000 people with their terrifying Latino underclass, instead of the boondocks. Who's more likely to shoot me?


Posted by: Cryptic ned | Link to this comment | 12- 7-13 12:51 PM
horizontal rule
142

124 is very good.

less months-to-years-long self-motivated deadlines

Huh, put that way, that's what I like about my current job (more like weeks-to-month rather than years and not totally self-motivated, of course, since our customers have a fair amount of input) for me personally, I have a personality that's inclined to mull over what I'm working on after I've gone home, it's nice to have work which benefits from that.

That said, I'm still really glad I didn't go into academia.


Posted by: NickS | Link to this comment | 12- 7-13 12:55 PM
horizontal rule
143

Trapnel even looks the part (for a 'tec rather than a street cop). And the Volcano could do with a hipster gumshoe among the clients.


Posted by: Alex | Link to this comment | 12- 7-13 1:13 PM
horizontal rule
144

Balloon Juice has a periodic thread for job searchers.

Honestly, I don't know what kind of job trapnel would like, and every thought that occurs to me is probably a case of projection on my part. That said: would you like to work in a progressive organization? Bill Moyers is all over the kinds of organizations working in that realm. And: what's your *subject matter*? Start there. Then look for organizations working in that subject matter.


Posted by: parsimon | Link to this comment | 12- 7-13 1:16 PM
horizontal rule
145

I hope that didn't sound glib. From what I can tell, trapnel would be an asset to any organization with a flexible organizational model. He thrives in a research environment. He likes a mission. That says to me that any number of organizations operating on the left would absolutely love to have him.


Posted by: parsimon | Link to this comment | 12- 7-13 1:22 PM
horizontal rule
146

He's a man with a mission. He could be a boy with a gun.


Posted by: Barry Freed | Link to this comment | 12- 7-13 1:31 PM
horizontal rule
147

I don't think trapnel is a boy, is the thing.


Posted by: parsimon | Link to this comment | 12- 7-13 1:40 PM
horizontal rule
148

That is, in case I haven't been clear, I think he's really smart. Anything that doesn't allow him to exercise that ability is going in the wrong direction.


Posted by: parsimon | Link to this comment | 12- 7-13 1:46 PM
horizontal rule
149

Lots of PDs are going high tech. Chicago's has a chief data officer. (Note: don't work for CPD.)


Posted by: ogged | Link to this comment | 12- 7-13 1:51 PM
horizontal rule
150

150: Brains seem like the sort of thing that might occasionally be useful in police work. I mean, maybe not consistently, but it could happen.


Posted by: LizardBreath | Link to this comment | 12- 7-13 2:06 PM
horizontal rule
151

||

Meanwhile, a cut of $8 to $9 billion in SNAP benefits, apparently supported by the Dem Senate Chair of the Agriculture Committee, is a really bad idea.

|>


Posted by: parsimon | Link to this comment | 12- 7-13 2:14 PM
horizontal rule
152

I don't think trapnel is a boy, is the thing.

I'm pretty sure he's a boy.


Posted by: nosflow | Link to this comment | 12- 7-13 2:23 PM
horizontal rule
153

Whatever you say, neb. Please don't call me a girl, though: I will be annoyed.

When does a boy become a man? O, it is a vexed question.

Vaguely related: I watched a film called Beasts of the Southern Wild last night, and my god, it was amazing. Highly, highly recommended. It is, on one level, about a girl growing up.


Posted by: parsimon | Link to this comment | 12- 7-13 2:29 PM
horizontal rule
154

Trapnel's nickname on the force could be "Brains."


Posted by: RH | Link to this comment | 12- 7-13 2:47 PM
horizontal rule
155

I suppose hardly anybody is old enough for "Dietrich" to catch on as a nickname.


Posted by: Moby Hick | Link to this comment | 12- 7-13 2:52 PM
horizontal rule
156

I'm incredibly conflict-averse.

I can easily imagine trapnel as the angelic half of a good cop / bad cop duo.


Posted by: torrey pine | Link to this comment | 12- 7-13 2:55 PM
horizontal rule
157

OT bleg: Does anyone here know how to solve an errors-in-variables multivariate linear regression? Or know where a solution can be quickly looked up online?


Posted by: Eggplant | Link to this comment | 12- 7-13 3:11 PM
horizontal rule
158

It seemed more tasteful to ask this on the smarm thread than the charity thread.


Posted by: Eggplant | Link to this comment | 12- 7-13 3:14 PM
horizontal rule
159

I bet Cosma does.


Posted by: nosflow | Link to this comment | 12- 7-13 3:18 PM
horizontal rule
160

Unless he has a grant due I'm going to have to find someone else.


Posted by: Eggplant | Link to this comment | 12- 7-13 3:22 PM
horizontal rule
161

Error in one predictor, or in more than one, or in the response variable?


Posted by: Kreskin | Link to this comment | 12- 7-13 3:25 PM
horizontal rule
162

159: I've never used those types of models, but I think it's just a form of latent variable whatnotery. And I'm not sure what you mean by "solve" here. One can look up the equations or the code fairly easily.


Posted by: Moby Hick | Link to this comment | 12- 7-13 3:28 PM
horizontal rule
163

Eggplant just found one of these at a yard sale.


Posted by: Kreskin | Link to this comment | 12- 7-13 3:31 PM
horizontal rule
164

163: My predictor is a vector, and there are errors in each component. Also errors in the response vector.
164: By latent variable whatnotery, are you referring to some method of moments estimator used as an instrumental variable? Also, where would I find the equations?


Posted by: Eggplant | Link to this comment | 12- 7-13 3:33 PM
horizontal rule
165

I'm assuming response vector is your dependent variable. Regular regression does errors in that.


Posted by: Moby Hick | Link to this comment | 12- 7-13 3:37 PM
horizontal rule
166

After a lot of work and failures my algorithm spit out some values that, while completely wrong, have a correlation with the accepted values of -0.94. Failure that total must mean something.


Posted by: Eggplant | Link to this comment | 12- 7-13 3:39 PM
horizontal rule
167

I'm not referring to a method of moments estimator. I'm not sure I've ever used that either.


Posted by: Moby Hick | Link to this comment | 12- 7-13 3:40 PM
horizontal rule
168

168: That's not failure. That's just inverted.


Posted by: Moby Hick | Link to this comment | 12- 7-13 3:41 PM
horizontal rule
169

167: Sure. I'm currently hoping my bad results are from errors in my independent variable, for which I've not correctly controlled.


Posted by: Eggplant | Link to this comment | 12- 7-13 3:41 PM
horizontal rule
170

170: Inverted and offset.


Posted by: Eggplant | Link to this comment | 12- 7-13 3:43 PM
horizontal rule
171

He thrives in a research environment.

I appreciate the thought, but there's really no evidence for this at all.


Posted by: x.trapnel | Link to this comment | 12- 7-13 3:49 PM
horizontal rule
172

For nearly everything I do, r = .94 would be an astoundingly strong correlation.


Posted by: Moby Hick | Link to this comment | 12- 7-13 3:50 PM
horizontal rule
173

174: Same here. I'm confused and excited.


Posted by: Eggplant | Link to this comment | 12- 7-13 3:54 PM
horizontal rule
174

I'm confused and excited.

Get a room, you two.


Posted by: essear | Link to this comment | 12- 7-13 3:56 PM
horizontal rule
175

If your predictor is a vector, why do you need multivariate regression?


Posted by: Moby Hick | Link to this comment | 12- 7-13 3:59 PM
horizontal rule
176

Because I misspoke just now. It's a set of vectors.


Posted by: Eggplant | Link to this comment | 12- 7-13 4:03 PM
horizontal rule
177

Did I mention I'm confused?


Posted by: Eggplant | Link to this comment | 12- 7-13 4:03 PM
horizontal rule
178

Maybe you should become a cop.


Posted by: teofilo | Link to this comment | 12- 7-13 4:15 PM
horizontal rule
179

Back when I checked Bay Area job listings, I saw a couple of data analysis jobs with SFPD. I'm sure there's a need for an app to track every chase, sorted by method.


Posted by: fake accent | Link to this comment | 12- 7-13 4:57 PM
horizontal rule
180

Are you using a statistical package?


Posted by: Moby Hick | Link to this comment | 12- 7-13 4:59 PM
horizontal rule
181

Both listings I'm thinking of were craigslist postings.


Posted by: fake accent | Link to this comment | 12- 7-13 4:59 PM
horizontal rule
182

Back when I was applying for planning jobs, I applied for one with the Ft. Worth PD. Sort of a data analysis thing with a spatial focus, IIRC.


Posted by: teofilo | Link to this comment | 12- 7-13 5:02 PM
horizontal rule
183

182: I'm not, but I could. At least, I could use R. I could get access to SPSS or newer versions of Matlab, but not without some bother.


Posted by: Eggplant | Link to this comment | 12- 7-13 5:19 PM
horizontal rule
184

Any of those should have a routine for what you need. You shouldn't need a more general latent-variable algorithm.


Posted by: Kreskin | Link to this comment | 12- 7-13 5:27 PM
horizontal rule
185

latent variable whatnotery

I'd always wondered what PROC LVW was for.


Posted by: Kreskin | Link to this comment | 12- 7-13 5:27 PM
horizontal rule
186

Apparently, SAS has a PROC CALIS especially for this. But using almost any package is about certain to be easier than programming your own matrix algebra.


Posted by: Moby Hick | Link to this comment | 12- 7-13 5:40 PM
horizontal rule
187

It's pretty easy to find documentation for Stata and SAS. The first useful link I find in R was by some guy.


Posted by: Moby Hick | Link to this comment | 12- 7-13 5:43 PM
horizontal rule
188

Reading back over this again, especially 171, I'm wondering why you decided to try this particular technique.


Posted by: Moby Hick | Link to this comment | 12- 7-13 5:48 PM
horizontal rule
189

Before going on, I'd get SPSS or R and try a standard regression. A -.94 correlation sound like it could be caused by something like not having a parameter for an intercept or some algebra-related typo.


Posted by: Moby Hick | Link to this comment | 12- 7-13 6:03 PM
horizontal rule
190

Are you using a statistical package?

Where it counts.


Posted by: Stanley | Link to this comment | 12- 7-13 6:19 PM
horizontal rule
191

I just thought of a really bad joke with "it's within the margin of error" as a punchline.


Posted by: Moby Hick | Link to this comment | 12- 7-13 6:41 PM
horizontal rule
192

I'm assuming PROC CALIS is like Cialis for statistical packages. If your calculations run for more than 4 hours, consult your doctor.


Posted by: fake accent | Link to this comment | 12- 7-13 7:08 PM
horizontal rule
193

190: Yeah, I wasn't very clear; beyond my normal levels of befuddlement I've not slept much the past few days. The error in the dependent variable is, as it turns out, significantly correlated with that of the independent variable. I first want to make sure I'm properly incorporating the errors-in-variables model, rather than relying on my crude method.
191: I think by not properly accounting for the correlation between dependent and independent errors I have effectively displaced the intercept term.


Posted by: Eggplant | Link to this comment | 12- 7-13 7:13 PM
horizontal rule
194

I'm glad Eggplant is getting answers and that his conversation thread is discouraging me from whining about relationship problems again. Maybe I should become a cop, except it's not really a good fit with the pacifism.


Posted by: Thorn | Link to this comment | 12- 7-13 7:41 PM
horizontal rule
195

I'm not sure where Eggplant is getting answers from. I'm more confused than when I started.


Posted by: Moby Hick | Link to this comment | 12- 7-13 7:56 PM
horizontal rule
196

I have a bunch of straight lines in a moderately high dimensional space. But for noise they would all intersect at one point. I would like to find that point. The noise additively effects the slopes of the lines and their "intercepts" (just some point the line passes through). This, I think, describes a multivariate linear regression with errors in variables. Right?


Posted by: Eggplant | Link to this comment | 12- 7-13 8:25 PM
horizontal rule
197

Back when I checked Bay Area job listings, I saw a couple of data analysis jobs with SFPD. I'm sure there's a need for an app to track every chase, sorted by method.

That's SF for you. They have a Chief Information Officer and are now supposed to be adding a concurrent Chief Data Officer.


Posted by: Minivet | Link to this comment | 12- 7-13 8:33 PM
horizontal rule
198

I'm not sure. I'm bad at story problems.


Posted by: Moby Hick | Link to this comment | 12- 7-13 8:33 PM
horizontal rule
199

They have a Chief Information Officer and are now supposed to be adding a concurrent Chief Data Officer.

What's the difference between these two positions supposed to be?


Posted by: teofilo | Link to this comment | 12- 7-13 8:36 PM
horizontal rule
200

It sounds right, so far as the error in measurement goes, but I'm not sure what the all intersecting at one point does to the regressions assumptions. Does a graph of the residuals against the predicted values look funky?


Posted by: Moby Hick | Link to this comment | 12- 7-13 8:57 PM
horizontal rule
201

201: It's probably just part of a hierarchy. The Chief Information Officer sits above the Chief Data Officer and the Chief Anecdote Officer.


Posted by: Moby Hick | Link to this comment | 12- 7-13 8:58 PM
horizontal rule
202

This, I think, describes a multivariate linear regression with errors in variables.

Why is this a regression problem? What are the independent and dependent variables?

I would think it's more just a minimization problem.


Posted by: essear | Link to this comment | 12- 7-13 9:00 PM
horizontal rule
203

Laydeez?


Posted by: Moby Hick | Link to this comment | 12- 7-13 9:10 PM
horizontal rule
204

Wait, that might be right. This may explain why my internet searches for solutions didn't turn up anything appropriate today.


Posted by: Eggplant | Link to this comment | 12- 7-13 9:21 PM
horizontal rule
205

Regression is a minimization problem, sort of. But it is true that I've never heard of anything like 198 solved with regression.


Posted by: Moby Hick | Link to this comment | 12- 7-13 9:22 PM
horizontal rule
206

Not "sort of". Just a specific type.


Posted by: Moby Hick | Link to this comment | 12- 7-13 9:22 PM
horizontal rule
207

The solution to the minimization problem ends up being pretty similar to ordinary least squares, but I guess that doesn't make it a regression.


Posted by: Eggplant | Link to this comment | 12- 7-13 9:23 PM
horizontal rule
208

Here's how I would approach it after a minute or so of thought, which is probably far from the optimal or most elegant thing to do.

If I just threw a bunch of lines at you, they're not in general going to intersect, but I could ask you to find the point where they come closest to intersecting. So you could take a point and compute its squared distance to each of the lines (the distance from a point to a line is a pretty easy thing to compute), then minimize the sum of the squares. Locally, if you go away from that the points get further away, so you can define a point with some kind of error ellipse.

If you have error bars on the lines, the first thing that comes to mind -- probably not the optimal thing -- is to just Monte Carlo the parameters of the lines, find the best-fit point with error ellipse for each random choice, keep a list of them all, and then try to combine them into some averaged answer in the end.

I'm sure there's something more clever you could do with the error bars, with a bit more thought.


Posted by: essear | Link to this comment | 12- 7-13 9:27 PM
horizontal rule
209

Oh! Or for a given point and a line with error bars, you could define a probability that the point lies on the line, in a fairly obvious way. Then you can do the usual likelihood-maximization thing.


Posted by: essear | Link to this comment | 12- 7-13 9:30 PM
horizontal rule
210

What about the probability that the point lies on the line unobtrusively?


Posted by: Moby Hick | Link to this comment | 12- 7-13 9:32 PM
horizontal rule
211

Unfortunately the error bars are unknown. Under the usual assumptions (independent, identically distributed) it's possible to get an algebraic answer, parametrized by the error.


Posted by: Eggplant | Link to this comment | 12- 7-13 9:34 PM
horizontal rule
212

The error bars are unknown? This just gets more and more confusing.


Posted by: essear | Link to this comment | 12- 7-13 9:37 PM
horizontal rule
213

But in that case it would seem minimizing the squared distance is the only thing one can do, no?


Posted by: essear | Link to this comment | 12- 7-13 9:38 PM
horizontal rule
214

Yeah, unknown. Also, the lines are nearly collinear, so everything is ridiculously sensitive.


Posted by: Eggplant | Link to this comment | 12- 7-13 9:42 PM
horizontal rule
215

This the worst live-blogging ever.


Posted by: Stanley | Link to this comment | 12- 7-13 9:44 PM
horizontal rule
216

The lines are nearly collinear and have unknown errors, but you're sure they intersect in precisely one point? I feel like this would make more sense if you explained where this problem came from. Or if I could think of a sufficiently outlandish story about how you're doing this as part of some highly secret government project to take over the internet or something.


Posted by: essear | Link to this comment | 12- 7-13 9:47 PM
horizontal rule
217

That's the second time someone has said almost those exact words to me.


Posted by: Eggplant | Link to this comment | 12- 7-13 9:49 PM
horizontal rule
218

219 to 217. I would take offense but I find myself pretty boring too.


Posted by: Eggplant | Link to this comment | 12- 7-13 9:52 PM
horizontal rule
219

My biggest adventure this week was putting shoes and a sweater on my dog.


Posted by: Eggplant | Link to this comment | 12- 7-13 9:54 PM
horizontal rule
220

And the lines are where the sweater should button.


Posted by: Moby Hick | Link to this comment | 12- 7-13 9:58 PM
horizontal rule
221

218: If you want to subject yourself to more of my writing wizardry I can email you.


Posted by: Eggplant | Link to this comment | 12- 7-13 10:02 PM
horizontal rule
222

150: Brains seem like the sort of thing that might occasionally be useful in police work. I mean, maybe not consistently, but it could happen.

That's just crazy talk.


Posted by: gswift | Link to this comment | 12- 7-13 10:08 PM
horizontal rule
223

putting shoes and a sweater on my dog

Do you have a collinear collie?


Posted by: fake accent | Link to this comment | 12- 7-13 10:21 PM
horizontal rule
224

A pitbull. He's normally quite gentle, but apparently really dislikes the stripes on his sweater; he has a linear aggression problem.


Posted by: Eggplant | Link to this comment | 12- 7-13 10:41 PM
horizontal rule
225

Now there's some liveblogging that Stanley should appreciate.


Posted by: Kreskin | Link to this comment | 12- 8-13 6:46 AM
horizontal rule
226

I enjoyed the liveblogging.


Posted by: Turgid Jacobian | Link to this comment | 12- 8-13 7:08 AM
horizontal rule
227

226-7: See, yeah, that's quality stuff right there.


Posted by: Stanley | Link to this comment | 12- 8-13 8:04 AM
horizontal rule
228

222 made me laugh out loud.


Posted by: heebie-geebie | Link to this comment | 12- 8-13 8:05 AM
horizontal rule
229

What's the difference between these two positions supposed to be?

The CIO will be responsible for the totality of the departments systems--for example, getting those computers in Johnny Law's car to work, and talk to the other computers; deciding where and how to store data collected by the department; and maybe managing the information collected by cameras, if the department has them. More of a tech/propellor-head type.

The CDO will focus solely on working with the data the department collects, and looking for patterns, trying to use predictive analysis, etc. Definitely a programmer/math type.


Posted by: ogged | Link to this comment | 12- 8-13 8:30 AM
horizontal rule
230

Buck is fond of pointing out that in the early 20th century, corporations frequently had CEOs -- Chief Electrification Officers. That stopped being a job as electricity got normal.


Posted by: LizardBreath | Link to this comment | 12- 8-13 8:36 AM
horizontal rule
231

I do think CIO as currently construed is a passing phase that will soon look as ridiculous as Electrification officers. Fucking business has been information processing, systems and flows forever, just increasingly augmented by computers and networks starting in the late 50s or so.


Posted by: JP Stormcrow | Link to this comment | 12- 8-13 8:49 AM
horizontal rule
232

Isn't it just a new title for prothonotary.


Posted by: Moby Hick | Link to this comment | 12- 8-13 9:03 AM
horizontal rule
233

To wrap up this morning, one can get the minimization problem into the form of a multivariate regression with errors in the independent variable, but I'm realizing those errors violate the assumptions required for the standard estimators to be unbiased. Also, the sweaters and shoes thing was pretty awesome.


Posted by: Eggplant | Link to this comment | 12- 8-13 9:55 AM
horizontal rule
234

I'm still not getting how this is a regression problem. Can you explain it to me?


Posted by: essear | Link to this comment | 12- 9-13 2:01 AM
horizontal rule
235

Well, it's not really, and I was probably using terms too loosely. Minimizing the squared distance gets an equation in the form of a multiple linear regression, so I was hoping to use the e-i-v models to handle my noise, but of course the noise doesn't obey those assumptions after all the algebra.


Posted by: Eggplant | Link to this comment | 12- 9-13 3:10 AM
horizontal rule
236

For the record, a lot of hospitals now have Chief Experience Officers (CXOs) due to all the reporting and incentive payments for patient experience levels.


Posted by: Minivet | Link to this comment | 12- 9-13 9:22 AM
horizontal rule
237

What's a patient experience level? How much they enjoyed their stay?


Posted by: LizardBreath | Link to this comment | 12- 9-13 9:24 AM
horizontal rule
238

239: The standard instrument is HCAHPS, which has a bunch of specific questions as well as a 0-10 overall rating.

Certainly there's going to be variation based on how well their stay turned out, and possibly based on other factors like how ritzy the hospital is (the wine-label effect). I'm not sure how well that's controlled for, but I know Medicare uses a lot of the specific questions (communication, responsiveness, management of pain) as well as overall rating, to determine value-based payments.


Posted by: Minivet | Link to this comment | 12- 9-13 9:49 AM
horizontal rule