So, one of the joys of Ubuntu is upgrading your OS every 6 months or so, getting kicked in the teeth, and spending the next week figuring out how to get back the functionality you used to have.

For me (this time) this meant figuring out why third-button emulation didn’t work on my two button mouse anymore (ancient, I know, but it ain’t broke).  This turns out to be’s fault, and to fix it I have to do the same thing I had to do to enable it 15 years ago when I was running FreeBSD — edit the X config file in vi.  It’s all covered here.  At least xorg.conf is not the godawful mess that XFree’s config file was.  Still, one of Ubuntu’s most consistent failings is the way graphical configuration utilities consistently lag the functionality they support.  Ubuntu switched to pulseaudio as the default sound system two full releases before a pulseaudio configurator was available.  This was a real problem since pulse frequently chose the wrong output device as the default, leaving users with no sound and no way to fix the problem except hand-hacking the pulse config file.

What really reminds me of pulseaudio, though, is Unity, Canonical’s new graphical shell.  I don’t want to be a “the new interface sucks because it’s not the old interface” guy, but Unity is not fully thought out and not ready to be the default desktop.  This guy has it pretty much right (although I don’t hate the new scrollbar).  I wanted to like it, but the mechanism for running applications is fundamentally broken, and that happens to be the main function of an OS.  I could go on for a while about how stupid I think it is for an OS to depend on 3D acceleration, but that just makes me sound like a curmudgeon. As the link points out, though, it’s self-evidently stupid to ship a 3D-only desktop if your OS doesn’t install 3D drivers by default because of its philosophy on proprietary software.

Readers (if there were any) might rightly point out that these things are my fault for upgrading a perfectly working system.  It turns out six months is just about long enough to decide that the things you had to go through last time weren’t so bad, and to convince yourself that the shiny new features you’ll get from updating will be totally worth it.

Now I’m off to figure out why xsane doesn’t work anymore.

On WikiLeaks

I’ve been reading as much about WikiLeaks’ release of the State Department cables as I can find, and trying to get a handle on what I and others think about it.  The reaction to these leaks by the media, politicians, and various public figures has been far more hostile than to WL’s previous releases.  The key question seems to be whether the release of the cables is a responsible action, in the manner of the classic whistle-blower, or whether it is egregious and unwarranted.

I think the hostility stems primarily from two causes, a misunderstanding of exactly what has taken place, and a bit of good old-fashioned American navel-gazing.

On the first point, I think most people are under the impression (as I was, based on the initial coverage), that WikiLeaks has dumped it’s entire cache of documents on the Internet for everyone to peruse.  In fact the documents are being released in redacted form by WikiLeaks’ media partners at a rate of about 100 per day.  As far as I can tell, the balance of the documents are not available anywhere on line, nor were they available on the old site.

My main point though, is aimed at the people who don’t think these documents are significant enough to warrant a deliberate breach of American security.  It’s only natural that these documents (being records of US diplomacy) are primarily of interest to people in other countries.  For instance, I think people in the UK are pretty interested to find out that their government lied to them about the cluster bomb treaty and the fate of Diego Garcia natives, and residents of the Middle East might want to know how the Qatari government manipulates al-Jazeera.  If someone sends you that kind of information, it seems to me that you’re morally obligated to disseminate it as widely as you can.

Perhaps the saddest part of this chapter is the vehemence with which the American media have attacked Assange and WikiLeaks, and the ambivalence they’ve shown in reporting the stories.  Glenn Greenwald has really been on top of this, check out the way the MSNBC host talks about Assange in the clip linked here.  Liberal media indeed.  Coverage in the New York Times, which has all the cables, has been, to me, significantly poorer than that of the Guardian, and even the English-language coverage by Der Spiegel.  Check out this list of stories broken by all of the outlets for a comparison (admittedly biased towards the Guardian).

I think the media has particularly dropped the ball on reporting the ideology behind WikiLeaks.  They have spelled out quite clearly the reasons behind what they’re doing, but I had to read about it on the blog of a grad student in California, because I haven’t heard it mentioned (much less engaged) by the mainstream press.  I suppose this is a little unfair, because the mainstream press does a bad job of reporting ideas and ideology at the best of times.

Why has the US media not embraced this treasure trove of information?  Maybe this is too simplistic, but I blame jealousy.  Once upon a time Daniel Ellsberg took the Pentagon Papers to the NYT, but the cables leaker(s) went to WikiLeaks instead.  Some potentially good reasons are spelled out in these videos.  This is yet another way in which new media is eating old media’s lunch, and a certain amount of hostility is only natural (if deeply unprofessional).

I’d like to close by saying that I don’t see a millimeter of moral difference between the releases by WikiLeaks and those of any other major media organization.  The reason they have been targeted is that they enjoy less status than more established organizations, and because they are a choke point in the information flow.  Next time you hear Eric Holder or some other Obama apparatchik call Julian Assange a criminal, just imagine they are talking about Marcus Brauchli or Bill Keller instead.  Better yet, ask yourself why those guys haven’t done anything to get themselves talked about that way.

Scientific Meta-Journalism

This post is almost two months old at this point, but I’ve just seen it. It’s a meta-article from the Guardian about science journalism, and it’s quite good. The biggest thing he misses is that the actual results of the study, which are probably quite technical and several steps removed from any general conclusion, will be fundamentally distorted to make them seem to have a massive impact on daily life.  This, however, might be done by the Journal’s PR people and not necessarily by the reporter.

The Global Game

A friend wondered, while watching Ghana v. Serbia on Saturday, how many countries in the 2010 World Cup have been bombed by the USA.  The ensuing discussion became led inevitably to this:

Countries bombed, shelled, invaded, occupied or attacked by the US:

  • Serbia (Kosovo War, 1999)
  • North Korea (Korean War)
  • South Korea (Korean War)
  • Japan (WW2)
  • Italy (WW2)
  • France (WW1/WW2)
  • Germany (WW1/WW2)
  • Mexico (Wilson administration and Mexican war)
  • Spain (Spanish-American war)
  • Algeria (Tripolitan war)

Countries where the US has organized, sponsored, or supported coups d’etat (not including the above):

  • Chile (Allende)
  • Ghana (Nkruma)

Countries where the US has supported brutal totalitarian regimes (not including the above):

  • South Africa (National party)
  • Argentina (Jorge Rafael Videla)
  • Greece (Colonels regime)
  • Paraguay (Stroessner)

That’s 16, or half of the teams in the Mondial.  I’ve probably also forgotten something obvious and embarrassing.

Honorable Mention goes to Honduras, whose banana republican governments were backed by the United States and the United Fruit for much of the 20th century, and to Portugal under the Estadio Novo, with which the US was allied through NATO.

It should be emphasized that if the US had not supported these regimes and organized these coups, the communists would have won, and we’d all be speaking Russian today.

Propaganda: because counting on apathy might not be enough

Propaganda because counting on apathy might not be enough

Propaganda because counting on apathy might not be enough

Inspired by



Top 10 Songs of Right Now

10. Matt and Kim – “Daylight

9. Lady GaGa – “Bad Romance

8. Nick Cave and the Bad Seeds – “John Finn’s Wife

7. Booker T and the MGs – “Stranger on the Shore

6. Johann Sebastian Bach – “Toccata and Fugue in D minor

5. The Antlers – “Two

4. Gaudi + Nusrat Fateh Ali Khan – Bethe Bethe Kese Kese

3. Elton John – “Someone Saved My Life Tonight

2. Air – “Talisman

1. Notorious B.I.G. – “Hypnotize

How many possible ways can the 2010 World Cup tournament unfold?

If you are filling out a bracket for the 2010 World Cup , here are the number of ways a bracket can go:

8 groups of 4 teams each. Top 2 move on, and the order is important. So that is 4 teams choose 2 teams or 6 ways to pick the two teams. Then, those 6 pairs of teams can be ordered 2 diff. ways, so 12 possibilities per group. There are 8 groups, so 12^8 possible ways to pick the top 2 teams in order from the 8 groups. Then, these 16 teams play a single elim tourney which is 15 games, so 2^15 possible ways that can go.

This gives 12^8 * 2^15 which is 14089640214528 (14 trillion or 1.4 x 10^13) ways for the tournament to go. Which of these ways do you hope or expect it to go? :)

(Btw, for comparison purposes, the NCAA basketball tournament with 64 teams has 2^63 or 9223372036854775808  (9 quintillion or 9.22 x 10^18 ) possibilities. )

Now, given the way the knockout stage works, you could have symmetries with respect to which teams play which. We don’t have to worry about those if we define our problem to also take into consideration the days and stadiums in which the teams play.

Good luck to all 32 teams!

A bit on Cell Phones and Cancer

I finally made time to read the results of the Interphone retrospective study on cell phones and brain cancer that were released last week.  What persuaded me to do it were the radically conflicting media interpretations of the announcement.  Here’s what seems to be the consensus view, as reported on the Washington Post’s “Post Tech” blog:

A large international study into the link between cellphone use and two kinds of brain cancer produced inconclusive results, according to a report to be released Tuesday in Geneva.

But researchers of the report noted flaws in the methodology of the long-awaited study. … The study’s results echo past research that the cellphone industry has cited for nearly two decades — a murky picture that there is not a conclusive link between cellphone use and cancer nor conclusive results that such a connection isn’t possible.

That’s the kind of summary I wish were written more often.  Ambiguity is probably the most common scientific result, although the press seems determined not to admit it in science reporting or elsewhere.  Contrast that with this take from, of all places, the BBC:

Analysis of more than 10,000 people by the International Agency for Research on Cancer (IARC) found no relationship between years of use and risk.  There is no known biological mechanism by which mobiles could cause cancer, but there has been public concern.  It is hoped this study will allay some anxieties, as research continues.

In other words: “You superstitious idiots only think phones cause cancer because you don’t know anything about Science.”  It’s one thing to say that no link was found, but to basically insult people who believe it could exist is pretty rich.  That first sentence is also carefully worded to focus on the one thing the study did rule out, instead of the many things it didn’t.  Logically one would expect the amount of use to be a driver of risk rather than length of use.  Why didn’t they lead with that?  Because the findings about the amount of use are inconsistent, and the way the two articles deal with them is telling.  One problem is that for most people there appears to be a risk-reduction associated with moderate, regular cell-phone use. Here’s the BBC’s take:

In fact, most regular users – defined as people who made use of their phone at least once a week – appeared to have a lower risk of brain cancer than those who rarely used a phone. The report authors stressed however this was unlikely to be down to any protective effect of phone use, and more a quirk of the study.

In the WaPo article this is seen as more than a mere quirk, indicative of a “flaw” and a “source of possible inaccuracies”.  The other key point is that people in the highest usage category (more than 30 minutes per day) did appear to show an increased risk.  Here’s how the WaPo handles that point:

The 10-year study, which was conducted by the World Health Organization’s International Agency for Research on Cancer, concluded there were “suggestions” that heavy use could increase the risk of glioma but “biases and error prevent a causal interpretation” that would directly blame cellphone radiation for the tumor.

According to the BBC story, these results were “dismissed as problematic.”  The Beeb story tries pretty hard to blame this on errors in reporting, citing the fact that ten people (out of over 5000 with tumours) claimed to have spent 12 hours per day on the phone.  They punch this further with a quotation from the report that the cancer “interferes with memory and cognition, undermining the accuracy of the recollections of such extreme use.”  That may be true, but what it adds up to is that you don’t know if that risk is there or not.  It doesn’t mean the risk should be discounted.

On the issue of funding for the study (about $24 million total), the BBC says that it “received some funding from the mobile industry”, while the Post tells us that it was “almost a quarter”.  Weirdest of all is the difference on further reports.  The BBC story seems to be pushing the idea that it’s a waste of taxpayer money, with this quote from the leader of the UK arm of the study:

“Whether it is worth doing more research, that is a question for society. These are expensive studies, and there are many other things in the world that should be investigated.  It is society which has to answer the question of how long you continue to investigate something that does not have a biological basis.”

He has clearly made up his mind.  According to the Post however, the lead researchers “urge more investigation into the topic to account for how cellphone use is affecting the health of youths” and because “the behavior of cellphone users has changed since the study was launched in 2000″.  That’s a bit different.  Either way, both articles report that a 30-year prospective study started last month.  So 30 years from now we ought to know, although that study too is funded in part by the cell phone industry.  If it gets canceled early, I might throw my phone away.

Free software and Free speech

I decided to start blogging here again because, although I’d hoped google wave would turn into a general interest forum, supplanting blogging, IM, and email, activity there has kind of petered out.  So although these thoughts are someone disjointed I figured I would put them here.

Although I had been singing the praises of SAGE for a while, and denigrating all non-open source mathematics packages…after coding the same project in both SAGE/python and Matlab professional (on a university system) I can say that I really see why people like Matlab so much.  About 70% of the time I spend in SAGE is in getting the various special data-types to work well together and finding the right way to invoke the interfaces to other projects.  Matlab takes away all the headaches by having so few data types to choose from and really simple and intuitive commands for loading and saving data.  Of course, I’m not ready to surrender to the proprietary software mongers, I just think that the creators of open source mathematics packages should aim to incorporate some of the desirable features of MATLAB for a mass-market audience.  Incomplete as it is, Octave does a pretty good job of emulating the basic functionality of MATLAB.  I would like to see and/or participate in work to extend the functionality of Octave so that it not only may catch up with MATLAB but also improve on it by incorporating some of the more algebraic aspects of computation that MATLAB is weak at (and that SAGE and Singular and other such packages were recently created in academia to address).

I was also thinking regarding the South Park/Mohammed episode: you would think that a lot of Muslims or Islamist-sympathizer-types with overseas insurgents would threaten right-wing talking heads when they go on news shows and call for various violent acts.  I guess they do, but they are never “censored”.  (side note: although a friend of mine was telling me in Ottawa that Ann Coulter had to cancel an appearance there because the administration said they couldn’t guarantee her safety and/or immunity from Canadian hate-speech law–which I think shameful for Canada as much of a mendacious clown as I think that commentator is…)  I think what got me thinking about this was Col. Lang pointing out on his blog that someone like him, who makes pronouncements, based on his scholarship, concerning Islamic society, could be next.  It’s really beyond my understanding why the issuers of this threat weren’t tracked down and questioned, so they’d know they were being watched, at least.  (According to Wikipedia: ] New York City Police Commissioner Raymond Kelly said he was aware of the website posting, but said, “We don’t think that this threat, as is currently assessed, rises to a crime right now.”[79])  Failing that response from the authorities, I like the attempt being made by some (such as Jon Stewart in particular) to talk about the episode and insult the issuers of the threat, so as to “defuse” it.  By making as many people as possible in society an “offender” against these fanatics sense of decency, (the idea would be) we can in effect make no one a particular target.
Two other comments about that Sic Semper Tyrannis comment thread above: I’m quite disappointed that, in the comments, Lang takes the position that assassination of a U.S. citizen abroad would be justified because he didn’t “turn himself in”.  Maybe we should’ve tried that against Marc Rich in Geneva or wherever he was!  Also, extending on the “diffusing the threat” theme, some commenter notes that “May 20th is ‘Everybody Draw Mohammed Day’.

Do Fruits and Vegetables Prevent Cancer or Not?

A friend linked this Times blog post today, and it made me seethe with rage.  The headline is “Eating Vegetables Doesn’t Stop Cancer,” which seems reasonable enough.  It’s based on a recent study in the Journal of the National Cancer Institute which, unfortunately, reaches the opposite conclusion.  The journalist is forced to admit, in the sixth paragraph, that there was a statistically significant reduction in risk – 1-4% for eating 200 grams/day above the average.  Ok that’s small, but it’s still the exact opposite of what was in the headline and the first five paragraphs.  How does she reconcile this?  Unconvincingly:

While the findings suggest at least a small lower risk of cancer among those who eat lots of vegetables, the slight difference could be explained by a number of variables, like reporting errors among the study subjects or the fact that vegetable eaters also are less likely to smoke or drink to excess. In addition, a 4 percent reduction in relative risk offers very little practical benefit to an individual.

Of course, reporting errors could also cut the other way.  I suspect people are much more likely to exaggerate the amount of fruit and vegetables that they eat, particularly those who eat relatively little.

There is news here, which is that the reduction is much lower than was previously hoped, as this editorial in the same journal makes clear.  This point is totally absent from the Times post.  My biggest problem would be that the overly reductive title implies that no fruit or vegetable confers any benefit against any cancer.  There is a lot of evidence that some do, which they mention briefly at the end.