Pondering the relative merits of staying in and watching tennis -v- going out and doing something in the lovely weather.

It’s not just the BBC, but they have the tv rights, so they can suffer my wrath.

It’s a shame, because tennis is one of the professional sports nearest to gender equality in terms of media coverage in this country. But all this talk of a Brit not having won Wimbledon since 1938 winds me up. Didn’t a Brit win in 1977?

Three more posts on the Iranian election wrt Benford’s Law; they don’t seem to think it indicates anything’s wrong.

And this provisional analysis (14 page pdf, and it reads like a blog in academic paper form) seems to conclude that election fraud may have taken place, but that then again it may not.

Benford’s Law anomalies in the 2009 Iranian presidential election — Boudewijn F. Roukema

Abstract: The results of the 2009 Iranian presidential election presented by the Iranian Ministry of the Interior (MOI) are analysed based on Benford’s Law and an empirical variant of Benford’s Law. The null hypothesis that the vote count distributions satisfy these distributions is rejected at a significance of p ≤ 0.007, based on the presence of 41 vote counts for candidate K that start with the digit 7, compared to an expected 21.2–22 occurrences expected for the null hypothesis. A less significant anomaly suggested by Benford’s Law could be interpreted as an overestimate of candidate A’s total vote count by several million votes. Possible signs of further anomalies are that the logarithmic vote count distributions of A, R, and K are positively skewed by 4.6, 5.8, and 2.5 standard errors in the skewness respectively, i.e. they are inconsistent with a log-normal distribution with p ∼ 4×10^-6, 7×10^-9, and 1.2×10^-2 respectively. M’s distribution is not significantly skewed.

It’s a 14 page academic pdf.

Questions for the statistically-capable lazyweb:

• Have people applied Benford’s Law to election results before?
• Is the empirical variant of Benford’s Law a standard technique, or something Roukema has made up?
• Do the stats calculations look legit? (I didn’t do that much stats, and I’ve forgotton it all anyway.)
  • (pp7-8) Why is assuming that the 36 samples are independent considered conservative?

Via Andrew Sullivan.

This week I am mostly hating everything to do with e-mail, although recovery CDs that don’t are running a close second.

TfL say

Oyster pay as you go will be accepted on all National Rail journeys within Greater London on Wednesday and Thursday, just show your Oyster card at station gate lines.

That’s not what the train operating companies are saying. For example, SouthEastern says

Southeastern has agreed to accept valid tickets for the Underground … on services that mirror Tube routes on our network. …

• Brixton to Victoria
• Elephant and Castle to Blackfriars / City Thameslink
• London Cannon Street / Waterloo East / Charing Cross

This BBC news story was earlier parroting the TfL line that oyster payg could be used on all national rail journeys within Greater London, but it’s now been updated to include this quote

National rail operators will accept Oyster cards on rail routes “that mirror Tube journeys” in Greater London on Wednesday and Thursday, a TfL spokesman said.

So why haven’t TfL corrected their website? Why didn’t TfL get it right on their website in the first place? What are they trying to do, sting everyone for penalty fares?

YouGov did a huge opinion poll just before the Euro elections (sample size: 32,268). The full report is a 15 page pdf full of tables of responses broken down by European voting intention (Con / Lab / LD / Green / UKIP / BNP). (The BNP got over twice as many votes as the SNP and Plaid Cymru combined, so YouGov won’t have got a large enough sample for SNP/PC.)

• Peter Kellner of YouGov wrote an article about it for Channel 4.
• Anthony Wells then wrote about it for his UK Polling Report site.

My thoughts:

• (p3) Wells says that BNP voters are disproportionately likely to come from Labour-voting families (47%, against 42% overall). More interesting to me is that while 42% of the country comes from a generally Labour-voting family, only 27% come from a generally Tory-voting family.

• (p6-7) BNP voters (and UKIP voters to a lesser extent) are much more likely to feel insecure in their everyday lives. The standout for me is that half of BNP voters don’t feel comfortable going out in their area, but they’re also more likely to feel pessimistic about their economic situation (which Kellner highlights).

• (p7) Apparantly white people are the most discriminated against group in the country. 40% think white people suffer unfair discrimination; only 19% think non-whites suffer unfair discrimination. (Kellner goes into more detail on this.)

  I know people are much more likely to notice discrimination they and their friends suffer than discrimination people they don’t know suffer, but even so.

• (p8-9) There’s a clear pecking order seen in journalism standards:

  • BBC journalists: TRUSTED by Green, LD, Lab, Con, UKIP; DISTRUSTED by BNP
  • broadsheet journalists: TRUSTED by Green, Con, LD; DISTRUSTED by BNP, Lab, UKIP
  • mid-market journalists: TRUSTED by no-one; DISTRUSTED by at least 2:1 for all parties
  • red top journalists: TRUSTED by no-one; DISTRUSTED by at least 5:1 for all parties

  But note how much less likely BNP supporters are to believe what the traditional media are telling them.

• (p10) ²/₅ of UKIP and BNP voters think the EU should set environmental policy! (As do clear majorities of the other parties’ voters.)

• (p14) It’s only 12%, but BNP voters are still much more likely to get their news from a political website than are other parties’ voters. (Which ties in with their greater distrust of traditional news media.)

I don’t think breaking up BNP press conferences is a good way of seizing the moral high ground. It makes anti-facsists look like violent, ignorant, elitist thugs, and gives the BNP credibility.

They have the same right to freedom of speech that we do.

I hope it won’t happen again.

It seems fashionable today to bemoan the BNP’s electoral gains, so I’ll add my 2p:

The BNP is not a problem. The BNP is a symptom of underlying problems.

I hadn’t realised it was supposed to be raining today.

My jsession id on the deathly slow intranet site I’m trying to use looks remarkably like “squishy doom”. Quite appropriate, given why I’m trying to use it.

(Actually, it looks like “squishy doom” with extra wailing added in the doom.)

By my reckoning, that’s three times England has provided three of the Champion’s League quarter-finalists, and twice that the non-English team has gone on to win it.

I have found a use for Twitter. Behold #eurovision (and boggle at the volume).

Is it in the spirit of Eurovision to watch with a cd of better music playing in the background?

I feel let down by some of my party’s MPs, some of them quite senior.

Remember what happens to ordinary people who inflate their expenses.

Just caught some spammers in the middle of a spam raid.

• two LJ comment notifications came in while I was checking my e-mail, both spam
• clicked through, deleted them both as spam; reloaded page to be sure: all spam gone
• switch back to my e-mail; delete offending comment notifications
• I have new e-mail!
• …is comment notification of more spam on same page

Hey LJ! If I delete 2 comments as spam within a minute of them being posted, and then another comment comes in on the same entry a few seconds later, maybe that’s spam too?

If DreamWidth gets better spam filtering, I’m more likely to switch over.

ETA: ...and then another 4 came in. And then 6 10 on this one.

$ perl -MFile::Slurp -e 'print "$_\n" for read_dir "."'
WC2B
t
.hg
.hgignore
$ perl -MFile::Slurp -e 'print "$_\n" for sort read_dir "."'
.

Add parantheses to taste. Continued fail. I have to add { $a cmp $b } to get it to work.

Sorry, I shouldn’t fill this thing up with computers and politics. I went to a lovely party on Saturday and got away with a lewd geeky joke in Janet+Owen’s birthday card. Then on the way back I eavesdropped on a drunken student’s entertaining conversation with a couple of acquaintances she’d randomly bumped into on the coach, and experienced the wonders of dawn on a speeding London bus that barely had to stop anywhere. Maybe I should travel at 5am more often.

I posted one tweet. And then… then I forgot my Twitter password.

I’ve started a petition asking for the police to be individually identifiable when on duty.

The policeman who assaulted Ian Tomlinson was wearing a balaclava and had concealed his badge number.

Please sign it, show it to people who might want to sign it, etc.

Wording partially copied from Rob argyraspid.