Tuesday, 30 August 2011

She Says

(another blues song)

My baby she threw me out on my ear
This story I tell you is sad
I turn round and say, "What you do that for?"
She tells me I've been bad

My baby says she loves me (My baby says she loves me)
My baby says she loves me (My baby says she loves me)
My baby says she loves me, but I been bad.

She says there's been all these women
She counts them one two three
But you knows they only for practice
They don't mean nothin to me

My baby says she loves me (My baby says she loves me)
My baby says she loves me (My baby says she loves me)
My baby knows she loves me, but I been bad.

She says that all my thinkin
Is driving her insane
My logic is ruining her awesome
Like pouring her lovin down the drain

My baby says she loves me (My baby says she loves me)
My baby says she loves me (My baby says she loves me)
My baby feels she loves me, but I been bad.

She said she would always love me
She said shed love me long
But you know she wouldn't even wait
For me to finish this stupid song

My baby says she loves me (My baby says she loves me)
My baby says she loves me (My baby says she loves me)
You know my baby loves me, but I been bad.

Well I guess I'm not the perfect man
I guess that's plain to see
But whose that guy whose been hangin around you?
Who the fuck is he?

My baby says she loves me (My baby says she loves me)
My baby says she loves me (My baby says she loves me)
My baby claims she loves me, but she been bad.

Monday, 29 August 2011

More Nymwars

The debate over real names in google+ continues. It's been going on long enough now for some actual information to emerge.

Tod Vierling has been collecting stats on the whole shebang. His results are updated fairly often, and if you think he's missed something, there is even a submission form. So far the overwhelming number of unique commentators are in favor of allowing pseudonyms in google+ and the majority of those use their real names themselves - thus: googles targeted audience would prefer a network where pseudonyms are allowed.

Eric Schmidt, (Google Exec. Chair.) has been recorded pointing out that;
G+ was built primarily as an identity service, so fundamentally, it depends on people using their real names if they're going to build future products that leverage that information

... and anyone who doesn't like that should get out.

This pretty much echos what I've been saying all allong: google does not want robust discussion or political activism on their network. They want a sort of boardroom etiquette, the sort of polite conversation you get at political fundraisers.

It gets worse: if you are a teenager wanting to avoid bullying (or pedophiles), or a woman who is not OK about sexual harassment online, they don't want you either. (Though they probably hope that the real names policy will discourage this behavior there is no evidence that it can.) Chris Bridges has a good summary

Basically this is the foot-gun approach to marketing. By denying pseudonyms, google are throwing away information (when people choose their own name, the name they choose tells you something about the person), and restricting the marketplace available to their customers (advertisers).

However, these revelations suggest a way forward: I submit that an identity verification scheme is a great addition to a social network. Eric is right when he says there are good uses for this and people who will want it and benefit from it. It is just should not be the basis for the network.

In addition to previous suggestions: allow people to use a pseudonym for their profile name, different pseudonyms for different circles, and keep a private "real name" between themselves and google, I'd suggest simply tagging profiles with a verification rating (those who are well behaved - the bad people get banned normally) something like:

  1. not reviewed: we have not checked this profile, so cannot say anything about it
  2. under review: looking into it (see below
  3. unconfirmed: profiles where the person is well behaved but we have not received satisfactory paperwork
  4. confirmed: pseudonymous profile, there is a real person behind it
  5. verified: this is a real person and, yes, this is their real name
  • may want to be able to mark profiles as removed or banned (keep their comments, or replace removed comments with some placeholder, so evidence remains to remind people and serve as a bad example).
  • allows for possible additional ratings such as google partner (gave us money) for commercial profiles.

Further: when a profile comes under review, it is handled like a bug report - a link to a review page, from the profile, which has a description of the problem and an opportunity for feedback. This will allow the development of some sort of due process without constitutionalizing the network.

I maintain that the above satisfied just about everyone's concerns and makes google+ a head-and-shoulders improvement on the other social networking offerings out there.

Thursday, 11 August 2011

Rice Pudding

Recipe: Rice Pudding in the Microwave (mostly).

This is my personal favorite method to make a quite nice pudding... by cheating.

You need butter, flour, milk, sugar and rice. None of the quantities will be exact... but roughly 50g, 1tblsp, 250ml, 1tblsp, and 1 cup respectively. It's the amount most cookbooks say would serve four - but if you try that you'll need ice-cream too.

Boil up the rice. It needs to be slightly overcooked so it's sticky but not gluggy. This is the longest part - the bit that makes it into a pudding only takes about five minutes. So if you do the rice normally, you can let it stand during the next bit, then drain, it should work out OK.

What we're gonna do is make a thick, sweet, white sauce and mix it in.

Using a microwave safe jug, I use a 2l pyrex jug, melt about 50g of butter. Needs to be completely liquid. Usually takes about 20-30secs. Stir in a tablespoon of flour. Any flour will work, I like to use self-raising flour: it makes things more fluffy. A rounded tablespoon is usually enough. The mixture should be wet but not runny.

Heat a cup of milk. That's a metric cup - 250ml. The milk should not boil, just warm. 30-40secs should be enough. The idea is that it shouldn't make the butter/flour mix go solid when you mix it in. If the milk gets a skin on top, it's too hot.

Gradually stir the milk into the jug. It won't mix evenly so don't worry.

Nuke for 2mins, then stir. A wire whisk is brilliant for this.

At this point you want a glossy, plasticky look. About the consistency of toothpaste.

Often it comes out too thick, that means too much flour. That's OK, just add more milk, a dribble at a time, and stir it in, until you get the right consistency. If it is too thin, don't worry right now, just go on with the next step. So you should err on the side of too runny.

Sometimes the mix has seperated. In this case, it will be liquid on top and almost solid under that. You'll need to break up the solid part and mix it in - squashing where needed. It can help if you interrupt the nukeage halfway to give it an extra stir. This isn't as bad as the lumps you get in gravey, they usually blend in. Even if they don't, the next nuke will get-em.

Stir in the sugar. It's a pudding so it has to be sweet. I use fine brown sugar for this.

Add flavouring if you want. Baileys is good, a dash of vanilla, caramel or butterscotch works well. This is usually needed since the regular method of making rice pudding in the oven lightly burns (caramelizes) the lactose in the milk, turning it into... caramel.

Now, if you added a mashed banana at this stage, you could portion it out and chill it overnight, and you'd have banana pudding. Add half a cup of your favorite cheese and you have a simple pasta topping (leave off the sugar). But this is rice pudding so drain that rice and stir it in.

Nuke for another 2mins and stir. If it wasn't too runny though, only go for another minute. You only need to cook in the flavor.

This time it should be a good consistency, holding it's shape when moulded. As before, if it is too thick, slowly add more milk. If too runny you have a problem: add flour, a level table-spoon at a time, stir it in, and re-nuke for a minute, until it's right or too thick.

You can keep it warming in the oven until it is needed. Portion up, garnish and serve.

Restaurants use an ice-cream scoop or some sort of mould. A 1-cup measuring cup works well. Then it can be served on a plate, and it looks good. People have been known to pour over a thin sauce like a glaze - caramel, chocolate, or a fruit sauce works well. Some pour a shot of brandy over it and set it on fire!

This is a very versatile and forgiving recipe, and, best of all, you can't burn it.

Tuesday, 9 August 2011


Managed to not stuff up on the speech, so I won. But not only won, managed to ace it! Now I represent the club at the regional competition on the 13th of September at the Telecom Conference Center. There I will be up against some very experienced speakers from the Auckland region.

And in the inbox - I have an audition tomorrow. The role is called "Underpants guy" - ooh, er...

Monday, 8 August 2011

Toastmasters - Humorous Speech Contest

... it's tonight: 5-7mins. Here's the draft:

I have a confession to make: I am a scientist. 
Before you judge: it isn't my fault! I tried hard not to become a scientist! 
I joined fringe religious cults.
I saluted the sun in the morning and danced naked in the moonlight. I partook of esoteric tantric rites - sometimes with company. 
I studied mystic crystal vibrations and alternative herbal medicines ... and that is the problem right there people: it's a slippery slope. 
It starts with the crystals and the herbs and it morphs grotesquely into geology, chemistry, botany and biology. All sciences! 
Seeking the middle path I studied perpetual motion and tried to get my over-unity engine supressed by the government but it was no good.  There was a last-ditch scrabble with pseudosciences, cold fusion, and flogistron but it was all over. The physics sucked me in! Dragging me down to the depraved depths of rational thought, critical analysis and skepticism! I must face the facts: I have become a science geek. 
For goodness sake: I have just spent the last three days working out the logistics of building a nucear power plant in my back yard! I need help! (Some of the bits are heavy.)

But I refuse to give it up! Though I am occasionally wrong... all right: frequently wrong ... I am addicted! It is a drug!

It is the lot of the scientist to be wrong. A lot. This is probably why women are under-represented. They just don't have the knack.
We have this big prize for the scientists who are the most wrong in each field. It's the Nobel Prize. 
Alfred Nobel worked out how to stabilise nitroglycerine. He figured this would be the end of warfare since the explosive was so powerful noone would dare to use it. This is an example of the special kind of wrong that scientists aspire to. He didn't get the Nobel Prize because, unlike politicians, when scientists invent an honor they do not automatically award it to themselves. 
Rutherford split the atom. He was a physicist, worked his whole life in physics. His major contributions were all to physics and he published in physics journals. So, naturally, he got the Nobel Prize for chemistry. He figured that the energy released in a nuclear disintegration was too small to ever have any practical use. This showed what a genius he was because this statement was so wrong that the opposite of it is also wrong - demonstrated by Oppenheimer, who also got the prize. 
The Nobel prize committee is usually stacked with women, and men being advised by women. It seems that even though women are bad at being wrong, they know it when they see it. Which leads to the question: if a man is wrong in the woods and there is no woman to hear him, how will he know? 
So please please please don't judge me too harshly - thank you.

Wednesday, 3 August 2011

Fission Man?

This just in:
Swedish police have detained a 31-year-old man in Ängelholm in western Sweden who was discovered after he sought advice from authorities on the legality of building a nuclear reactor in a domestic kitchen. 
Yeah right!
The man, who explained that his interest in nuclear physics was awakened as a teenager, ordered some radioactive material from overseas and acquired more by taking apart a domestic fire alarm.
Smoke detectors use americium 241, in the form of americium oxide - a typical smoke detector contains of the order of a milligram of the stuff. The critical mass for Am241 is 60-80kgs. Nice try.

It sounds like he was just piling fissionables up and hoping something would happen.

You could presumably make a sub-critical reactor with Am241, if you had an energetic particle source... say a big van-der-graaf generator and you use the proton stream. Not done the math. You can use Am242 to make a kind of battery.

He felt he wouldn't have been able to use it as a power source because:
To get it to generate electricity you would need a turbine and a generator and that is very difficult to build yourself.
... actually, those are the easy bits. It would fail to be a power source because it couldn't possibly have worked. The police let him go.

This guy was just building a box with some cool metal in it. But he reminds me of this guy building a cruise missile in his back yard. Again, over-sensationalized: a better description would be "a model jet aircraft". It would also have kept him out of trouble but he was trying to make a point.

Bottom line: you can bud all kinds of things if you are prepared to stretch a point.

Monday, 1 August 2011

Pseudonyms update

+Bradley Horowitz talks about the whole pseudonym thing.
Still can't get past the thing that Google does not want google+ to be the sort of place where people are anonymous or known by silly names. Like it detracts from the tone of the place I guess.

I can see how having people called "God" about can do that. However, not all pseudonyms are silly. Obvious pseudonyms are better (more honest) than plausible ones - which is what the policy basically asks for.

By restricting to real names, Google+ could never become a force for social and political change as we have seen with facebook for fear that some extremest opponent may track you down. Could never become a place for robust mature conversation. Just the sort of polite non-discussions you get at office parties or political fundraisers. Is that really the idea?

I maintain that the no-fake-names policy will keep good people away while doing nothing to discourage the bad ones.

Since other fields can be used to store and display aliases - why not do it the other way around: allow the profile name to be anything non-silly, and put the real names in other fields?

This could even be expanded to allow different identities in different circles - eg. someone could be Benny to family, Benjamin to "business colleagues", and Mr Hill to "current students". (And Honey-buns to lovers.)

Wouldn't you love this?