Posts by David Hood
Last ←Newer Page 1 2 3 4 5 Older→ First
-
I was up later than I really should have been working out how to make this graph
-
Hard News: Terror panics and the war imperative, in reply to Farmer Green,
No, that is the polite way of saying "We don't know what is going on but have our doubts"
This is the polite way of saying scam in general reading
http://www.science20.com/a_quantum_diaries_survivor/cold_fusion_a_better_study_on_the_infamous_ecat-146700And this is the detailed way of saying scam in technical circles
http://arxiv.org/ftp/arxiv/papers/1306/1306.6364.pdfBut it is really not worth spending too much time over. If you have someone convicted of fraud claiming to have built a machine that contradicts modern physics, the tests should not be taking place under conditions they dictate with them handling all the equipment.
Actually running it without it being connected to external power sources would also be a good thing, but the experimental design is another matter entirely.
-
Keith, Basically nothing much changes unless Labour got a bigger share of the vote- about a 3%ish increase nationwide, at which point New Zealand First would have the balance of power. I'm talking 3%ish increase in Labour without other parties raw numbers changing, rather than a swing.- so the people who stayed home this time coming to vote.
-
I realise it has been a couple of weeks since numbers were being bandied around in this thread, but here is my election contribution, showing my working.
-
Speaker: Why we should not dismiss…, in reply to Nick Melchior,
It might be helpful to have some examples of conspiracy theories that do make sense
http://spectrum.ieee.org/geek-life/history/the-great-lightbulb-conspiracy
-
I'm only just settling down to pick through the election results, but it is clear Kiwi's understand MMP. The electorate winner in that respect is Ōhāriu, outdoing even Epsom, as at a minimum half the voters must have split their vote (the theoretical maximum for all electorates is mostly near 100% but is much more tricky to calculate as an upper bound).
At the other end of the minimum spliters scale from Ōhāriu and Epsom (followed by Waiariki) we have Rodney and Manurewa (followed by Tauranga) -
Speaker: An Open Letter To David Cunliffe, in reply to Stephen Judd,
Some, but it is all basically provisional until specials are in, since the '11 totals include specials. Like most people who play with data, I will probably be busy the weekend after final results are in. Look for something after then. What I have done, though, is gone through the Wayback machine and gathered up the '11 (approx.) electorate level enrollment figures from the electoral commission pages that no longer exist (as they only publish the up to date ones)
https://www.dropbox.com/s/71yhgul5ukkj5wc/electoraldata11.txt?dl=0 (utf-8 text file)
While the '14 figures are available (so people could gather them a little easier), here they are gathered up in the same form.
https://www.dropbox.com/s/lj79xbyyppx0fx3/electoraldata14.txt?dl=0
This is how I am normalising the results on the basis of the number of voters in each electorate.
-
Speaker: An Open Letter To David Cunliffe, in reply to Paul Williams,
Trevor’s lost some ground in Hutt South, but that’s partly boundary changes and his relative decline is still a lot better than many others I suspect.
Nope, while boundary changes might be the cause, Hutt South's decline (depending on how you measure it, and I am measuring it as percentage of Labour vote (pre specials) to electorate size at election time relative to percentage of Labour vote to electorate size at previous election) was in the worst half-dozen or so.
Not that I think if it is time for Mallard to think about retiring has much bearing on the leadership issue.
-
For what it's worth, the change in Party Vote as percentage of people on the Electoral Roll (2014 vs 2011, 2014 before specials) in Ilam is entirely consistent with the general trend for Labour. And with an overall trend line that strong, I don't think it's local issues affecting turnout much.
-
Just reading about Jarrod Gilbert over in the dirty politics thread, I'd like to say I would be quite happy to put my name into a website run by a trusted party that assigned me a OIA request that someone who didn't want the government tippid off by who was asking wanted forwarded onto the government. In such an imaginary website I could check and veto a request before it went out under my name (as I wouldn't be wanting it to be used for sending threats etc) and because I would, in an imaginary way, be using an email address assigned by the site the response could be automatically fowarded by the site.
Now the idea needs a little refinement, most obviously around the authorising stuff sent out under my name and maybe some limits on who it can be sent to (govt addresses?) but I thought I would throw it out there, because while I know a lot about data matching I also know a lot about things that would confound data matching.
