"To express ernest disapproval" of something (Random House); to avert by prayer; from "de" + "precari", to pray.
The appearance of "deprecate" in the software-engineering community (about 25 years ago ISTR, give or take a decade) seemed magical: a solidly classical -- even slightly arcane -- word popping up in a powerfully appropriate place ("for security reasons, use of this feature is now deprecated") in the middle of a community that was notoriously indifferent to linguistic finesse. (On a whimsical study in 2003, I found that the ratio of incorrect to correct spellings of "supersede" was higher on web pages that also contained the word "megahertz". [If you're wondering, the anti-megahertz is charybdis, whose appearance on a web page almost guarantees that any supersede on that page will be correctly spelled.])
http://www.genecards.org/cgi-bin/carddisp.pl?gene=ADRA2B&snp=99 lists 99 SNPs in the ADRA2B gene. The following 8 of these SNPs also appear in my 23andMe results:
rs3813662
rs29000571
rs4907299
rs2229169
rs29000569
rs4426564
rs28932482
rs35053873
.
The available abstract doesn't say where on the gene the deletion mutation occurs, so I can't tell which of these SNPs is closest to it. (I would expect that a deletion mutation would never be a SNP.) Since they're all pretty close together, any one of them would be very likely to track the mutation over many generations.
NASA has one expert stating that the flyby of asteroid 2012 DA14 is a once-in-forty-years event, and another expert stating that this Russian meteor/meteorite is a once-in-100-years event, and yet another expert asserting that the fact that these two events happened on the same day is a coincidence. Given that the standard statistician's definition of "highly significant" corresponds to a 1% chance of coincidence, and that one day out of 40 years is 0.007%, I think the probability that NASA has spoken hastily is greater than the probability that this was a coincidence.
Can't we just agree to ignore all announcements about quantum computers that aren't accompanied by both (a) the system reference manual pages for the instructions that manipulate the quantum hardware, and (b) performance numbers for a completely specified problem?
The headline claims a revelation of "Opposition to Science", but the post itself only makes assertions about climate skepticism. If the writer really equates the two, he should try to calm down.
Why shouldn't I disable *every* root cert that I'd be surprised to find authenticating a browser session with my bank? Only a handful of root certificates suffice for nearly all of my secure browsing. Wouldn't I very much want to be alerted if an SSL session with my bank were based on a certificate issued by the Lower Slobbovian Postal Authority?
I hope this incident leads browser makers to adopt more realistic certificate husbandry mechanisms, such as alerting me when my bank's cert changes.
Let's eyeball the statistics: 60 measurements per organ (according to the abstract), times seven organs (adrenal glands, brain, gonads, heart, kidneys, liver, and spleen), times two feeding durations (5 and 14 weeks), times two sexes (male, female), time three strains of corn tested (NK 603, MON 810, MON 863), equals 5040 measurement series. (Wow! 7!) So in the absence of any effect, we should expect 5040/20 = 252 "statistically significant" (p <.05) discoveries and 5040/100 = 50 "statistically highly significant" (p <.01) discoveries.
Are we learning something about health, or are we just illustrating the perils of data dredging?
"Their genomic organization was strange and a little unexpected," says Batzer. "It appeared much more bird- and reptile-like than mammalian, even though it is indeed classified as a mammal."
Batzer is one of the researchers. He should know that in the standard evolutionary phylogenetic model, the platypus is no more closely related to birds than we are, since we and platypodes share a common ancestor that existed in the mammalian line after the mammalian line separated from birds.
While it is true that to a human a platypus looks more like a bird than does another human, it is equally true that to a platypus a human looks more like a bird than does another platypus.
And the implication that classifying it as a mammal is a mistake ("even though it is indeed classified as a mammal", rather than "even though it's a mammal") is just silly.
You can see some original work from these researchers at
http://staff.bath.ac.uk/pssmjb/finger/reportonstaf fdata.htm .
Unfortunately, the level of statistical literacy is disappointing. First, they observe that the mean digit ratios for men and women are 0.98 and 1.00 (respectively) without mentioning the standard deviations, an omission that makes the numbers almost meaningless. Then they "prove" the validity of their self-measurement system by boasting that its results are strongly correlated (r=0.819, p<0.001, "extremely significant indicating good inter-rater reliability"), as if a strong proof that two measurements are not uncorrelated were what we wanted, rather than a measure of the discrepancy between them. At that point, I stopped reading, despite the delicious political incorrectness of their work.
Here are the results from a study (some years ago) of numbers of web pages containing correct and incorrect spellings of "supersede" as a function of other words appearing in the same pages:....... hits with.......
other word "supersede" "supercede" ratio
The bias intended was not always achieved. "Murine" was supposed to sample zoologists, whom I think of as being classical eggheads, but pulled in a lot of molecular biologists, who are more like engineers, whose devastating effect is shown by the result with "megahertz". "Virgil" was supposed to get classicists, but got a lot of guys named that. The uncannily effective "charybdis", for reasons obvious in retrospect, produced a lot of legal briefs. Maybe lawyers are good for something after all.
The last two, "accommodate" and "accomodate", pursued the thought that correct spellings of "supersede" might be correlated with correct spellings of another commonly misspelled word. There is a correlation, but it's not clear that this technique is worth the added effort. Note, however, that someone who *misspells* "accommodate" is still more likely to get "supersede" right than is someone who *correctly* spells "megahertz".
I'm free to filter my mail any way I want. That's good.
A bunch of spam fighters who want to pool their statistics to fight spam more efficiently are free to do that. That's good.
If they want to put their results online where everybody can see them, that's good.
If my ISP wants to improve my email experience by letting me use the spam fighters' online database, that's good. (They do. I like it.)
If an ISP thinks most of their customers wouldn't want to wrestle with the details of spam filtering, and would prefer to have the ISP make the filtering decisions, that might not be the service I want, but I'll defend their right to offer that service.
If SBC has rented my daughter a damaged IP address, one with a bad reputation for spamming, that's bad: I nearly missed a message from her that went into my Spam folder. I hope my daughter can switch to a service that tries harder to keep its IP inventory clean. Absent that threat, I don't see why SBC would make any effort to reduce spamming.
So, SBC is free to decide how vigorously to discourage spamming. My daughter is free to choose her ISP. I am free to use the information that spam-fighting groups share freely. Life could be far worse.
But my daughter, having innocently rented this damaged IP address, must be having trouble sending email to people who filter their email with blocklists. That's bad. But aside from the spammers, I don't see whose freedom we should abridge to alleviate the problem.
Decide whether you want to manage
on
Geeks in Management?
·
· Score: 2, Insightful
You've received a lot of good advice on how to be a good manager, but first you must decide for yourself whether you want to be a manager. Some people just aren't born to manage: I was made a manager 25 years ago, and had to quit to get back to the photosynthetic rung of the food chain, where I've been ever since with the exception of 6 gruesome and rewarding-to-nobody months about 6 years ago. So don't let this important decision be made for you by accidents like nobody really suitable being available.
If you decide to move to the world of pie charts and performance evaluations, take the change seriously: you're learning a whole new job, and it will take study and attention.
Slashdot Mirror
"To express ernest disapproval" of something (Random House); to avert by prayer; from "de" + "precari", to pray. The appearance of "deprecate" in the software-engineering community (about 25 years ago ISTR, give or take a decade) seemed magical: a solidly classical -- even slightly arcane -- word popping up in a powerfully appropriate place ("for security reasons, use of this feature is now deprecated") in the middle of a community that was notoriously indifferent to linguistic finesse. (On a whimsical study in 2003, I found that the ratio of incorrect to correct spellings of "supersede" was higher on web pages that also contained the word "megahertz". [If you're wondering, the anti-megahertz is charybdis, whose appearance on a web page almost guarantees that any supersede on that page will be correctly spelled.])
". . . to deprecate iptables . . ."? I do not think "deprecate" means what you think it means.
http://www.genecards.org/cgi-bin/carddisp.pl?gene=ADRA2B&snp=99 lists 99 SNPs in the ADRA2B gene. The following 8 of these SNPs also appear in my 23andMe results: rs3813662 rs29000571 rs4907299 rs2229169 rs29000569 rs4426564 rs28932482 rs35053873 . The available abstract doesn't say where on the gene the deletion mutation occurs, so I can't tell which of these SNPs is closest to it. (I would expect that a deletion mutation would never be a SNP.) Since they're all pretty close together, any one of them would be very likely to track the mutation over many generations.
NASA has one expert stating that the flyby of asteroid 2012 DA14 is a once-in-forty-years event, and another expert stating that this Russian meteor/meteorite is a once-in-100-years event, and yet another expert asserting that the fact that these two events happened on the same day is a coincidence. Given that the standard statistician's definition of "highly significant" corresponds to a 1% chance of coincidence, and that one day out of 40 years is 0.007%, I think the probability that NASA has spoken hastily is greater than the probability that this was a coincidence.
Can't we just agree to ignore all announcements about quantum computers that aren't accompanied by both (a) the system reference manual pages for the instructions that manipulate the quantum hardware, and (b) performance numbers for a completely specified problem?
The headline claims a revelation of "Opposition to Science", but the post itself only makes assertions about climate skepticism. If the writer really equates the two, he should try to calm down.
I hope this incident leads browser makers to adopt more realistic certificate husbandry mechanisms, such as alerting me when my bank's cert changes.
Are we learning something about health, or are we just illustrating the perils of data dredging?
Batzer is one of the researchers. He should know that in the standard evolutionary phylogenetic model, the platypus is no more closely related to birds than we are, since we and platypodes share a common ancestor that existed in the mammalian line after the mammalian line separated from birds.
While it is true that to a human a platypus looks more like a bird than does another human, it is equally true that to a platypus a human looks more like a bird than does another platypus.
And the implication that classifying it as a mammal is a mistake ("even though it is indeed classified as a mammal", rather than "even though it's a mammal") is just silly.
You can see some original work from these researchers at http://staff.bath.ac.uk/pssmjb/finger/reportonstaf fdata.htm .
Unfortunately, the level of statistical literacy is disappointing. First, they observe that the mean digit ratios for men and women are 0.98 and 1.00 (respectively) without mentioning the standard deviations, an omission that makes the numbers almost meaningless. Then they "prove" the validity of their self-measurement system by boasting that its results are strongly correlated (r=0.819, p<0.001, "extremely significant indicating good inter-rater reliability"), as if a strong proof that two measurements are not uncorrelated were what we wanted, rather than a measure of the discrepancy between them. At that point, I stopped reading, despite the delicious political incorrectness of their work.
Here are the results from a study (some years ....... hits with .......
ago) of numbers of web pages containing
correct and incorrect spellings of
"supersede" as a function of other words
appearing in the same pages:
other word "supersede" "supercede" ratio
(none) 380,000 115,000 3.3
erudite 569 179 3.2
erudition 600 124 4.8
epistemological 1300 255 5.1
etruscan 147 25 5.9
murine 205 47 4.4
megahertz 549 340 1.6
virgil 1130 285 4.0
hellenic 892 154 5.8
morphogenesis 80 32 2.5
charybdis 260 21 12.4
accommodate 37,300 8,690 4.3
accomodate 1,200 626 1.9
The bias intended was not always
achieved. "Murine" was supposed to sample
zoologists, whom I think of as being
classical eggheads, but pulled in a lot of
molecular biologists, who are more like
engineers, whose devastating effect is shown
by the result with "megahertz". "Virgil" was
supposed to get classicists, but got a lot of
guys named that. The uncannily effective
"charybdis", for reasons obvious in
retrospect, produced a lot of legal
briefs. Maybe lawyers are good for something
after all.
The last two, "accommodate" and "accomodate",
pursued the thought that correct spellings of
"supersede" might be correlated with correct
spellings of another commonly misspelled
word. There is a correlation, but it's not
clear that this technique is worth the added
effort. Note, however, that someone who
*misspells* "accommodate" is still more
likely to get "supersede" right than is
someone who *correctly* spells "megahertz".
I'm free to filter my mail any way I want. That's good.
A bunch of spam fighters who want to pool their statistics to fight spam more efficiently are free to do that. That's good.
If they want to put their results online where everybody can see them, that's good.
If my ISP wants to improve my email experience by letting me use the spam fighters' online database, that's good. (They do. I like it.)
If an ISP thinks most of their customers wouldn't want to wrestle with the details of spam filtering, and would prefer to have the ISP make the filtering decisions, that might not be the service I want, but I'll defend their right to offer that service.
If SBC has rented my daughter a damaged IP address, one with a bad reputation for spamming, that's bad: I nearly missed a message from her that went into my Spam folder. I hope my daughter can switch to a service that tries harder to keep its IP inventory clean. Absent that threat, I don't see why SBC would make any effort to reduce spamming.
So, SBC is free to decide how vigorously to discourage spamming. My daughter is free to choose her ISP. I am free to use the information that spam-fighting groups share freely. Life could be far worse.
But my daughter, having innocently rented this damaged IP address, must be having trouble sending email to people who filter their email with blocklists. That's bad. But aside from the spammers, I don't see whose freedom we should abridge to alleviate the problem.
If you decide to move to the world of pie charts and performance evaluations, take the change seriously: you're learning a whole new job, and it will take study and attention.