Posts by Idiot Savant
Last ←Newer Page 1 2 3 4 5 Older→ First
-
Garth: yes - a year or so ago, at around midnight on Tuesdays.
Under no circumstances should you watch 4 without watching 3. It's a novel (one big, 5 series novel), and like Freemon says back in S1, "all the pieces matter".
The DVD is available overseas now, or you can always resort to the usual methods.
(And Americans only now seem to have got State of Play. Which is impressing them as much as The Wire. Huzzah!)
-
Season four of The Wire starts on March 10 at 12.25am on a Monday night. WTF?
It's good TV - "the best show on television" according to a fair chunk of critics - therefore it must be completely buried in a timeslot where no-one is awake.
And TVNZ wonder why people turn to the internet...
-
Craig: In an MMP environment, voter coalition preferences matter. They provide useful data to both parties (who want to know how not to piss off their supporters) and to other voters (who may very well change their voting preferences depending on their desired outcoming and assessment of which way a party will swing). So, I'd very much like to see more information about this. I'd just like it to be good, robust data rather than bullshit based on the opinions of a handful of people.
Sadly, I really can't see the media doing it. You need a sample size of ~400 to get an acceptable margin of error, and at current support rates you'd ned to call tens of thousands of people (assuming they all respond) to find your sample. And good luck finding 400 random United Future or ACT voters - they're so thinly spread hat you're needing ~100,000 respondents (at which stage you're polling a substantial portion of the electorate. At huge expense).
You could I suppose glom the data together from repeated polls, but then time and changing opinions will likely make the result meaningless.
-
Deborah: you may be assuming they read past the executive summary there.
-
Also, that particular editorial and the Armstrong column confirm (at least to my mind) that the Herald gets most of its information these days from blogs. The telling comments "admittedly based on a tiny number of respondents" & "Admittedly, the samples of minor party supporters are tiny" to my mind can only come from reading Idiot/Savants analysis on norightturn.
Or by talking to Audrey Young. She at least had the decency to be embarassed about her overinterpretation of data. Armstrong and whoever writes the editorials keeps peddlign it anyway though.
And the numbers are slightly worse than suggested - as a reader pointed out to me yesterday, some of the percentages are a giveaway. For NZ First, there were 15 or 16 respondents, but 90% to 9.1% means 9 to 1, so only 11 of them expressed a coalition preference (and 10 for National). For the Maori Party, it's 11 respondents, and 4 to 3 for National. For UF and ACT, as we already know, it's three people.
This is the sort of data which would make a social scientist embarassed (I know social scientists are the butt of many jokes, but one thing you can credit them with is basic stats). Sadly, journalists seem to have no sense of shame about using data sane people would regard as junk.
OTOH, if the polling company (who ought to know these things) doesn't tell them which bits are meaningful and which aren't, can we really blame the journalists for reporting what they're fed?
-
Simon: I've only been blogging for 5 years, not 15.
-
I think he pretty much had all the funny with the words "party vote for Act" really.
But Roger Douglas is back with them. Don't people love him anymore?
-
Having said that though, talk to a few Maori public servants about the way in which the foreshore and seabed protests (and the in-effect ban on Maori public servants having a voice on that issue) were handled if you want an example from the other end of the spectrum
I'm aware of those examples - one of someone being dismissed for her support of the Maori Party - and I did my nut about them at the time.
-
I really think this is extremely improper behaviour from McCully -- and a direct attack on my livelihood.
Did you expect any different froma party which is threatening public servants to discourage them from applying for the position of State Services Commissioner?
Unfortunately, having lain down with the kiwiblog right, they seem to be institutionalising thuggery as a political tactic.
-
I think one of the problem with talking about concensus is that it is practically impossible to get scientists to _absolutely_ agree about anything. It is in fact their job to question stuff and say "Maybe, but I'd really like to actually see the raw data on this".
Unfortunately this can also be a problem when trying to get scientists to talk to the media. You'll never get a scientist to call anything "safe" or "definite". There's no such thing as safe. There is always a risk even if it is 0.00001% likely to ever occur. But this refusal to label anything safe can lead to panic "Medicine/Food/Bicycle/Process type X is UNSAFE!!!!"
Absolutely. Science is about organised, systematic doubt, so naturally scientists hedge. They're never 100% certain, they always want more evidence, and they're always looking for holes in the theory to be examined. These are Good Things, and they're why science can produce exceptionally robust conclusions, such as those we see on anthropogenic climate change.
As for comparing climate change deniers to Holocaust Deniers, I think its entirely fair, as they display the same level of intellectual dishonesty and complete disregard for the facts.