Ideas and Solutions

MacBook Pro Retina or MacBook Air not sleeping when closing the lid?

2013-09-25T15:41:00.000+02:00

This only seems to happen to the MacBook Air and MacBook Pro Retina. Why?
I think I finally have the answer: They do not have built-in network ports.
Most people use them connected to both power, and a wired network port, through an external adapter when working at the office.
When leaving for the day, you close the screen, unplug the power, the network, and leave. Tomorrow your battery has died, and you cold boot. But only sometimes. Why?
In System Preferences, Energy Saver, Power Adapter it has a setting to "Wake for Network access".

So, after closing the lid and going to sleep:
If you unplug the power cable first, then the network, your MacBook stays asleep.
If you unplug the network cable first, then the power cable, your MacBook wakes for the network event, and never receives a new sleep instruction, as the lid stays closed.
Case closed.

I am too poor to be able to afford a loan from Wonga

2013-04-09T17:30:00.000+02:00

Wonga offers loans at ridiculous rates. I am too poor to throw money away like this:

The total cost works out to 74% for one day!
The tip says: "We suggest you choose your next pay day as the repayment date". If they really cared about their customers, I can think of several more relevant tips, such as:
"Hey, we noticed you chose the minimum amount for 1 day. Did you know that keeping it for 3 days won't cost you more?"

Yes, if you borrow more it becomes relatively cheaper, but now you know how they can afford prime time advertising on DSTV. I wish my shares gave this kind of a return. Currently Bitcoin seems to keep up, but who knows where that will end?

Advertising Mathematics

2013-02-18T12:57:00.000+02:00

I suspect that radio advertisements on a national station such as 5fm must be pretty expensive.
Certainly, if I had to pay for such an ad, I would probably check the maths before giving the go ahead. Apparently people in advertising, copywriting and editing do not seem to think so, though. (Unfortunately I cannot recall which companies were actually advertising; shows you how effective the ads were, or how defective my memory is). Two recent examples (paraphrased, but the numbers are correct):

According to BigInsurer the average time spent waiting for a call centre call to be answered is 38s. [Does that include the time spent listening to and navigating the IVR menu? I doubt that...]. This means that you spend 1.2 years of your life waiting for a call centre agent. [1.2 years ≈ 38 000 000 s, which seems to imply that the average person makes a million calls to call centres over their lifetime. I hope to make significantly fewer calls...]

According to SomeCompany, your heart will beat 3 million times in your lifetime, so make them count. [Indeed, make them count! At 60 beats per minute (don't get excited or exercise) that is just more than a month (34.7 days). I do agree with the fact that your life tends to stop just after your heart stops beating, if you can't get it going again soon...]

Or maybe it is a filter, to ensure that they don't target mathematically literate people?

Power blackouts

2013-01-10T16:26:00.001+02:00

I just checked my Landis+Gyr prepaid meter's hidden counter (i 050) and see that I have experienced 39 power failures since the installation of the meter (20 months ago).

Wow. That averages 2 power failures a month. Thanks Eskom!

Customer loyalty programs

2013-01-10T13:53:00.000+02:00

Everyone has a customer loyalty program these days, encouraging you to earn points in exchange for allowing them to track exactly what you buy, and when you buy it, and to get you to use them exclusively.

However, I would like to single out a few I have noticed these days that are not worth it: Edgars used to have a Cash Card, which gave you R50 back for every R700 you spent, which was a very respectable 7% discount, and made me buy things from them I normally would have bought elsewhere (like an Xbox from CNA, rather than Game - same price, but 7% discount!). This has now been replaced with a Thank U card, which works out to less than 1% discount. No thanks, I'll shop somewhere else then.

Similarly, Exclusive Books has a Fanatics card that rewards you with about 2% (after they charged you 20% more than Kalahari.com for the physical book and 200% more than Amazon.com for the ebook). No thanks.

Woolworths has the right idea though - 10% (I have seen 20% on some items) discount on specific items, deducted immediately at the till point! No fuss, no waiting for vouchers, and I know exactly what I am in for!

Banks, Credit Cards and Status

2012-10-25T10:43:00.000+02:00

Traditionally you start out with the normal credit card, for poor plebs.

After working hard and earning more money, you one day "graduate" to the gold card, with higher limits, more status, benefits (and fees).

When you start earning the big bucks, you finally qualify for the platinum card, with stratospheric limits, oodles of status, benefits (and even higher fees).

I'm just wondering, now that the gold price has truly and properly overtaken the platinum price, and gold costs more (per ounce, not per volume - yet), are the banks going to rebrand? Or can the middle class with their gold cards silently laugh at the so-called rich with their cheaper platinum cards?

Gullibility and the R1 cappuccino

2012-10-11T08:27:00.000+02:00

I heard a radio program the other day about things you know about your job that no-one else knows. One caller stated that he knew that the cappuccino you pay between R12 and R16 for only costs R1 to make, and everyone was horrified about the profit margins and accepted the claim at face value. He is in the industry, so he should know, right?

This is where critical thinking, as opposed to gullibility comes in. Let's assume a standard cappuccino. That consists of 250ml milk (give or take), 1 teaspoon sugar (4 grams), and about 7 grams of ground coffee beans. The exact amounts can vary, but let's assume that is accurate for a first estimate. That results in costs of R2 for the milk, almost nothing for the sugar, and about R0.80 for the coffee. The paper cup and plastic lid are not free either - call it another R0.20 at least. We are up to R3 for a cup already, and even if we made a mistake with some of the prices, we are never going to reduce that to below R1.

Of course you need to pay the barrista, pay for the machine (lease/buy, cleaning, repairs, various consumables) and rent the location to sell the coffee from.

While I do not deny that large profit margins can be made from selling cappuccinos, they are most definitely not on the order of 1000%, but probably between 100% and 200%.

What this exercise illustrates though is how few people are able to apply a bit of critical thinking and some quick mental arithmetic to blanket statements made by "experts". No wonder people keep falling for pyramid schemes and offers that are too good to be true.

How to get rid of junk mail

2012-10-10T08:16:00.000+02:00

The real, physical kind, not email (still working on that one...)

Reader's Digest is a master at sending you voluminous junk mail that fills your mail box. I might have won the sweepstakes dozens of times already, not to mention prompt reply cars, quads and limited lifetime offers.

If you ever actually reply (even to say no thanks, but enter me in the sweepstakes anyway, since no purchase is necessary terms and conditions legal fine print etc.) you will severely increase the number of offers extended to you, since you have just indicated that you are interested, as soon as they find the right offer to tempt you with.

What does dampen their enthusiasm, though, is if you just send the entire envelope, unopened, back to sender. They can't get you to buy anything if you don't open it, and now they have to pay postage on the entire package twice (not just the reply-paid envelope inside). And their processing centres are not set up to handle this.

This means that you are flagged as a "BAD PROSPECT" and left alone for a while. I have successfully reduced my spam from them from about two offers per month to about two offers per year.

Good luck!

Cellphone operators and crime (8ta, CellC, MTN, Vodacom)

2012-10-09T17:31:00.000+02:00

Your cellphone has been stolen. I hope they can't make phone calls...

It was switched off. You never changed the default pin. Good thing the default is:

8ta:
0000 or 1234 - no thief will ever guess that right? Your fraudulent call costs may be more than 0...

CellC:
A random number you couldn't remember, and changed to your birthday/lucky number/whatever. The thief won't get it in 10 tries before the SIM is blocked.

MTN:
00000 - no thief will ever guess that right? Your fraudulent call costs may be more than 0...

Vodacom:
A random number you couldn't remember, and changed to your birthday/lucky number/whatever. The thief won't get it in 10 tries before the SIM is blocked.

Now, blacklist it (otherwise the insurance won't pay, and it is the right thing to do, right?)

I'll use Startpage to do a search for the term "Blacklist" on each of the network operator's sites to see if they provide instructions. (Powered by Google, but without the filtering/sorting)

Let's see how our major network operators do when trying to blacklist a cellphone:

8ta:
The first non-paid for link has the number (081 182):

CellC:
The first non-paid for link has the number (084 140):

MTN:
Nothing...

Vodacom:
The first non-paid for link has the number (082 111):

Draw your own conclusions...

PS: The number for MTN is 083-1-808. It can be found on the MTN service provider page (mtnsp.co.za) or the SAPS. Good luck actually getting the phone blacklisted if complaints on Hello Peter are anything to go by.

Cellphone SIM lock and RICA

2012-10-09T16:49:00.000+02:00

We have had RICA (Regulation of Interception of Communication Act) for more than a year now. I don't see cellphone crime and theft coming down. Probably because RICA was never designed with a proper goal, but rather just another knee-jerk reaction. (Or maybe it is achieving exactly what was intended - keeping tabs on law-abiding citizens, but that is another discussion)

Why are Multichoice DSTV decoders not stolen? They are worth quite a bit more than some cellphones.... Simply because they are linked to a Smart Card and cannot be linked to a new card (at least not easily). So steal one, and the DSTV service disappears, with no way of getting it back or reselling the decoder.

Imagine if RICA worked like that! Every cellphone sold will have a SIM linked to the IMEI number (automatically, and enforced by law). If you really, really need to get a new SIM, you need to jump through the current RICA registration hoops, but because less than 1% of subscribers would actually need it, the checks can be properly implemented, and easily policed.

Steal a cellphone, and unless you can change the IMEI number (possible, but harder than replacing a SIM), or plan to export it (also a slightly less immediate way to turn into cash) it is essentially worthless.

Maybe this will prevent people getting killed for cellphone?

GPS coordinates - why S26.5 is not the same as S26 50'

2011-09-28T18:19:00.000+02:00

Everyone and his dog now has a GPS, and very often businesses helpfully post their GPS coordinates so that potential customers can find them. The problem comes in, however, when people do not realize that there are 60 seconds in a minute, and 60 minutes in a degree, and that depending on the format used in the particular GPS device, might be displayed in one of several formats.

Lets look at the coordinates for the London Bridge:
51° 30′ 29″ N, 0° 5′ 16″ W or 51° 30.4833 N, 0° 5.2666 W (Garmin default) or 51.508056, -0.087778

View Larger Map
If you mistakenly confused the formats, and posted the link as
51° 30.29 N, 0° 5.16 W:

View Larger Map

Fortunately the even worse error of:
51° 50′ 80″ N, 0° 8′ 77″ W is clearly wrong, and in such a case the original can be guessed.

In short, just as 0.5 hours = 30 minutes, 0.5 degrees = 30 minutes.
Safe navigating!

Anti-innovation companies

2011-09-07T09:17:00.000+02:00

Some companies, such as Google and Apple are all about innovation, and competing with them is hard because they keep moving the goalposts. This is part of why they make so much money and are so valuable to investors. In order to get this innovation, you need to actively solicit new ideas from staff, and properly incentivize innovation and ensure that you do not let great ideas slip through the cracks.

We know about these companies, and their stories are not new. Today's topic is so called "anti-innovation" companies - companies that place so many hurdles in the way of innovation and have so much red tape that getting a new idea through is almost impossible - you need to fortuitously combine the person with the idea, the person with the mandate, the person with the budget, the person with the resources (staff) and the correct phase of the moon, and also accept that the credit will go to some totally unrelated person to get a new idea implemented. People with innovative ideas unfailingly leave the employ of these companies after a while, completely disillusioned and disheartened, having accomplished nothing. Sometimes they will eventually see their idea implemented a decade or so later, when everyone else is doing it, and bestowing no competitive advantage whatsoever to their original company (and of course, no credit to the originator).

Examples are of course governments, big banks and insurance companies, and probably a large percentage of companies where there are more than 10 000 employees and the average employee has worked for the same company for more than 5 years - i.e. a large bureaucracy and stagnating, trapped employees. Promotions are only available when your immediate supervisor retires or dies (or sometimes gets promoted himself). The position tends to go to the person with the longest service, not the most qualified.

These anti-innovation companies are also big customers of managing consultants (McKinsey, Accenture), big audit firms and research firms. They spend millions on these firms to tell them what their employees have been telling them for years. Sometimes they even read the final report they have spent millions on.

If you recognize your employer, and like to innovate, leave. If you like a steady paycheck and get your stimulation elsewhere, chill. Whatever you do, do not buy shares in this company, and if you have, sell them. You can do better elsewhere. They may not go out of business, but they will also never shoot out the lights.

And so things will remain, until one day, Google or Apple (or some other company that innovates) decides that they want a slice of the revenue pie where our anti-innovation company operates - then they are dead in the water and will cease to exist - how long will it be before you replace your bank with Google Money or iBank if they became available? If an organization with as many inefficiencies and as much bureaucracy as your bank can make money, how long do you think it will be before an innovative company decides to target a share of that revenue? (EBay's Paypal is going into offline payments as we speak).

If Multichoice sold DSTV like Telkom sells ADSL...

2011-06-21T09:25:00.000+02:00

First, keep in mind that both companies are supposedly trying to maximise profit...

Applying for a DSTV installation would result in a date more than a month away, instead of today or tomorrow if no physical equipment is installed
If your decoder broke, tough. Buy another one. No-one would fix it.
Even if you had a dish and decoder, getting connected would take at least 2 weeks.
If DSTV ran out of satellite bandwidth, you would have to wait a year or so for a new one to be launched. No pre-planning would happen.
You would be charged R150 p.m. for the privilege of having DSTV, before even being allowed to view any channel.
You would then pay extra for whatever you want to view, per minute.
International content would cost a lot more, and be capped to 3 hours per month. Watching more would cost R2 per minute.

Now. if Telkom sold ADSL like Multichoice sells DSTV:

An independent technician would able to install a physical line today or tomorrow, if you didn't have one.
There would be a swap-out service for routers, for a nominal fee per month.
If you already had a line and router, getting connected would be a matter of minutes.
Telkom would not run out of ADSL ports, ever - they would plan ahead based on demand and unused ports, and not wait until after the last port has been used, and the next customer cannot be accommodated to begin ordering new equipment.
You would not be forced to pay R150 for the privilege of having a landline even if you didn't want to make calls.
Your ADSL package would only be capped at entry level, and the top package would not only be super fast, but uncapped as well.

If only...

Seatbelts vs Crime

2011-05-31T10:47:00.000+02:00

In South Africa, the big problem is violent crime, and everyone is very concerned.

This can be used to demonstrate how people misjudge risk, and make bad trade-offs: If I could invent a device which would reduce your chance of being killed in a criminal act by half (statistically proven), I would probably be a billionaire in a very short time, as people would be willing to pay thousands for such a device, and probably be willing to fork out a monthly subscription fee as well. They would force their children to get one, and if someone owned such a device, and didn't use it, they would be labeled as lunatics and gambling with their lives. If they neglected getting their children to use it, they would be ostracised in polite society, and would be ashamed to admit it.

Now, in South Africa, our road death toll is roughly equivalent to our crime death toll. A device exists that reduces your chance of being killed in an accident by much more than half - and comes standard with every single car, for decades already. It is called a seatbelt. Yet by my informal survey, between a third and two thirds of people do not wear their seatbelts regularly; some even fasten them over the empty seat and sit on them to silence the warning buzzer. The figures for children are actually worse - very few people would hesitate to take a quick trip with their children not in a car seat or wearing a seatbelt. This includes smart, educated people with the financial wherewithal to afford car seats...

The number one cause of unnatural death in children is car accidents, where they often have not been properly restrained...

And in case you feel like ignoring this because you do not stay in South Africa, keep in mind that in almost all other countries the chances of dying in a car accident is at least an order of a magnitude higher than the chance of dying due to a criminal act, yet I am willing to bet you would still consider the Criminal Death Preventer....

If you are one of the people religiously wearing a seatbelt, and doing the same for your children, you probably agree with me 100%.

Email - what an unfortunate name

2011-05-25T11:26:00.000+02:00

This morning I heard about thousands of Sarah Palin's emails being used in a new book, and her discontent over the matter. This once again set me thinking about the misnomer "email" and all the icons and preconceived ideas we live with.

Warning to the technically inclined - I am oversimplifying here, but you are welcome to read the actual spec for the message format RFC2822 or SMTP protocol RFC2821 for a more detailed overview.

Lets first look at mail, not email. We have a sender, and a recipient. If they are located in the same country, normally only one carrier is involved (that country's postal service) that takes responsibility for the delivery. If the sender and recipient are in different countries, there would normally be only two carriers involved - the sender's carrier, and the recipient's carrier, with a single handover. Both the sender and the receiver normally know exactly which carriers are involved. If the mail is in an envelope, tampering or interception is detectable by the recipient, unless expertly done (and such tampering or interception is normally prohibited by law, with rather strict penalties). It is almost impossible to do a bulk intercept, and in general you have a reasonable expectation of privacy.

In the electronic world, we have the rather unfortunate case of email, where the name suggests an electronic version of the manual process (RFC2822 even refers to an envelope and contents), which seems to imply that all (or most) of the assumptions for mail still holds for email.

In actual fact, and I think this is what is missed by most people, there is no "envelope" in email that wraps the "message" and protects it from prying eyes. In this aspect, calling email epostcards would have been more appropriate, as the message is clearly visible to anyone handling it, without leaving any traces, regardless of the expertise (or lack thereof) of the interceptor...

Even worse, there can be many carriers involved, and it is normally not possible to predict which ones will be involved. This allows for bulk interception and tampering by anyone on the mail path. This gives email its resilience, since the sender's carrier and recipient's carrier do not have to be directly connected, and the failure of any nodes on the path would normally not prevent delivery (the wonders of TCP/IP...)

There are also generally no laws protecting email with anything even close to the rigor of normal mail. Deleting an email is also not equivalent to burning a normal piece of mail, as quite a few copies would still be floating around...

In short then, expect very little privacy in your email communications, and if you need more, use PGP.

In fact, I think this blog is being read by fewer people than my email :-)

(And yes, I know that this is not how Sarah Palin's emails came to light. At least not this time.)

More information on the Landis+Gyr cashpower 3 phase prepaid electricity meter

2011-05-20T14:59:00.000+02:00

I have now figured out code 055, and have updated the original post to reflect that. With that knowledge comes some more information:

Dates and times are stored not in the traditional computer science epoch format, as that would require 10 digits, and the display only has 8, but in a slightly modified format: Dates and times are stored to a resolution of 1 minute, as the number of minutes elapsed since midnight on 1/1/1993.

Thus, the current date/time according to the meter would be 9667217 (2011/05/20 08:17). New vouchers contain 20 digits, and must contain at least the date/time of purchase, value, and some check digits or encryption, and possibly some link to the unit's serial number. The value has a precision of 0.1 kWh, and it is not inconceivable to purchase more than 10 000 kWh at once, requiring at least 6 digits. So, for neatness, lets assume 8 digits for the date/time and 8 digits for the value, leaving 4 digits for a checksum, and the 16 digits of significance are possibly XORed with the serial number, repeated twice.

Or, the maximum purchase could be 9999.9 kWh, requiring 5 digits, and the last 7 digits of the serial number could be used as well, but I am leaning towards the 8/8/4 theory, as the design would undoubtedly have required single digit errors to be detected.

More on this later as I figure it out...

All of this is nicely documented in IEC 62055-41 and 62055-51, including how to convert a 20 digit voucher to the 66 bit token, etc. I might get the standards to help me at some later stage. Anyway, it seems like cracking token won't be simple, but the standards might shed some light on control tokens that might allow some more cool things...

RIAA and MPAA vs drug companies - something smells rotten

2011-05-19T13:25:00.000+02:00

Watching Snow White while reading about drug companies and their expensive research, something hit me.

What is worth more to humanity - a new miracle drug that could cure HIV/AIDS or another great Disney movie? Stupid question, I know. Which is more expensive to make? In case you didn't know, researching a new drug could cost $500 million or more; most new movies cost less than $50 million, so new drugs easily cost 10 times more than new movies.

A pharmaceutical company is allowed a 20 year patent on a new drug (Forbes) from registration (before any trials - it could still take up to a decade before it gets to market); yet Snow White is still under copyright 74 years after release?

Generic companies are hailed as saviours when then copy a drug, and make a cheap generic available, yet a 12-year old gets arrested for copying a single song?

Am I the only one who thinks that something is rotten if this is allowed to go on?

Fixing home loan rate (2)

2011-05-17T07:56:00.000+02:00

Quick one:
Seems that Moneyweb finally got with the program, and are essentially saying what I did yesterday, in this article, just in more words, and less clear.
Go figure. (Maybe they read this blog - I wish!)

Sony PSN security and usability fail (After 15 May 2011)

2011-05-16T12:27:00.000+02:00

Yes, this is not directly about the famous PSN hack, but rather some more comments on how their security and usability still suck.

<rant>
When I could finally log back on, I had to change my password with a link sent to my email address. OK, this makes sense, as they could not trust their passwords anymore. But why does my password have to look like this?

Your password must:
- contain at least eight characters
- contain at least one letter and one number
- not contain any character used more than two times in a row
- not match the sign-in ID

Which means that Pasword1 is valid, yet YouWillNeverGuessThisPassword*&%^$%#%^&^ is not?
Nor is oInYhgbYNCyhsdUBuo valid, or AnyPassPhraseThatHappensToRepeatACharacterSomewhere173 or AnyPasPhraseThatHapensToMisOutOnANumber.

Also, if you click on any menu item, and try to cancel, you need to confirm that you want to cancel. The last time something this bad was invented was when you had to click on Start to Shut Down.
</rant>

Should you fix your home loan rate?

2011-05-16T08:48:00.000+02:00

Currently all the talk in South Africa is about your home loan rate, (Prime at 9%) and whether or not you should fix this rate, given that rates could rise by up to 2% in the next 2 years... Moneyweb article

The answer is actually simple - do you think you know more than the economists at the banks, and are you willing to bet on it? If not, keep your variable rate...

Let me explain. If you currently have a mortgage at prime less 1.5% (7.5%) variable rate, you will not be able to fix it at 7.5%. You will be charged up to 2% more for the "privilege" of fixing your rate.

This means that if you are afraid that rates could go up over the next two years, banks offer you the option of paying the higher rate now, to assuage your fears, and remove the uncertainty of having the lower rate for a longer time.

What you should do, if you cannot afford to take a chance that the rates go up by more than 2% in the next two years? Simple. Get a quote from your bank at what the payment would be if you fixed your interest rate. Now pay this amount monthly, without fixing your rate. You will be building up a nest egg, as you are in effect reducing the outstanding capital faster. Should the rates go up by less than 2%, you will still keep paying the same amount; the extra amount will just shrink. If the rates go up by more than 2%, and you cannot afford the new payment, ask your bank to recalculate your loan, using the extra capital to reduce the monthly payment.

Here is an example based on a R1 million loan at prime less 1.5%(7.5%) over 20 years (payment R8055.93 pm). Fixing the rate at 9.5% will result in a new payment of R9321.31 pm. Lets assume that the rates go up by 0.5% every six months over the next two years, and you keep the rate variable but pay the new amount. After two years you will have reduced the outstanding capital by an extra R17749.36, and even if that rate climbs by another 0.5% to 10%, you should still be able to afford the payment after recalculating to factor in the reduced capital (R9354.99).

You only lose if the rates go up by more than about 4% in the next two years. If the banks' economists thought that a likely outcome, they would have charged you more than 2% extra to fix your rate...

Why Twitter is not suitable for research and deep thoughts

2011-05-12T09:07:00.000+02:00

Blogs are still better for the communication of research ideas and deep thoughts, for several reasons.

First of all, tweets are too short to transfer meaningful amounts of information relating to research or thought. Secondly, due to their instant nature, people take less care with the grammar and spelling. Editing tweets and refining their content also never happens.

This all conspires to make tweets more like gossip. Here is what this blog would look like on Twitter:
@Joburg So u think my meter is faulty and I have to get a prepaid metee? WTF #billingcrisis

My new #prepaid meter installed today. Finally! Took them long enough #billingcrisis

Seems like there are some hidden codes. Please RT link to site explaining them if someone has it

OK, found a few codes for #prepaid meter. Might add them to site late - too many for tweet.

WTF is up wit #Colgate #toothpaste ? Tubes are smaller, but the price is still the same? #ripoff

F***ing #Taxis! #rant

Someone I know just got R500 fine + 1 point for stopping at a red robot! #WTF #AARTO

Just read review of new BlueMotion #Polo. LOL comments - clearly fuel consumption progress and safety not well understood. Just Google it - should find something.

F***ing #taxis and #idiots on the road! #roadrage

Just caught another #Mythbusters episode. Busted #deathray for the third time. Maybe they should use soldiers next time?

Anyway, that is why you won't find my thoughts on Twitter. Maybe one day I'll be smart enough to transmit all I need in only 140 characters. Until then, this is how I'll share my thoughts and ideas...

Mythbusters and Archimedes' solar death ray

2011-05-07T08:47:00.004+02:00

This was busted three times: in episodes 16, 46 and 157. However, I would like to revisit it, and look at one variable they may have missed. Let us look at the facts:

Mirrors (even bronze ones) reflect the sunlight.
Reflected sunlight from enough sources can cause the temperature to go up (quite significantly)
Precision is required to aim more than one mirror at the same spot.

I think that the first two points have been conclusively proven more than once, including by the mythbusters. The problem seems to be the last point, and they tried to address it the last time, but I think they still missed out on one variable - soldiers.

The shields were not used by students or scholars in a fun experiment, but by trained soldiers. Give 500 scholars a longbow, without training or practice, and two hours later you might also conclude that they are useless as weapons of war, as aiming is problematic and their arms get tired. A result which has been conclusively disproven in quite a few wars...

Get a 100 soldiers, and teach them to aim their mirrors properly for a few weeks (and also to keep them impeccably clean). Replace the ones that fail until you have an elite force of 100 soldiers that are death ray sharpshooters. Repeat the experiment.

I think that would change the status to "plausible..."

Road rage and driver education

2011-05-06T13:39:00.000+02:00

How often have you waved your hands at another driver and swore at them in traffic? And sometimes it escalates into full blown road rage, and someone gets hurt (or worse). In order to see the benefits, lets turn this question around:

How often have you improved your driving behaviour due to someone else swearing at you and waving their hands in traffic? Oh, sorry, you were already a perfect driver (or at least above average, like the rest of us, lol), but humour me... Right, not even once.

So, if you have never improved your driving based on the free education offered by others, what makes you think that the opposite will hold true?

Stop swearing at others, and concentrate on your own driving, and at avoiding the people who makes mistakes instead. The roads might not be safer, but at least you will be calmer. (And use less fuel, according to the Mythbusters See the last myth on the exploding steak episode...)

Fuel consumption today vs two decades ago

2011-05-05T11:19:00.000+02:00

It is almost impossible for any online car review to mention the miserly fuel consumption of their latest steed without some illiterate person making disparaging comments, implying that their Toyota/Opel/VW of twenty years ago was almost as good anyway. (See Polo Blue Motion review for an example).

What seems to be invariably forgotten/ignored in the comparison is that their old car weighed 30%-60% less - reason being the added safety equipment today. Crash that old car at 60km/h and you are guaranteed severe injuries, if not death, whereas the modern, 4 or 5 star Euro NCAP car would let you off almost unscathed. Assuming, of course, that you have the - surprisingly rare - ability to buckle the seat belt. No wonder airlines feel the need to still educate us on their use today...

Another culprit is the SI unit for fuel consumption - l/100km. A consumption of 5.1 does not sound much better than 6.3, whereas it is, in fact, a significant improvement, that can be further confused by mathematics: If car A has a consumption of 5.1 l/100km compared to car B with 6.3 l/100km, car A's consumption is 19% less than car B, but car B consumes almost 24% more fuel than car A.

Lastly, the consumption claims are always compared "on the open road" - where we spend very little time. Compare those identical cars in congested traffic, and a different picture will emerge.

I happen to have the figures for a few cars from 1998 (only 13 years ago, but still proving my point). The modern competitors have been chose to match power and acceleration, not engine size. The figures for 1998 are from an auto motor und sport article (1998) and the 2011 figures are from the german car websites.

Car	0-100 km/h	Power	ECE/EU urban	ECE/EU open road	% increase from open road to urban	Euro NCAP
1998 Alfa Romeo 156 2.5 V6 24V	7.7s	140kW	16.5 l/100km	8.4 l/100km	96%	N/A (YouTube videos look bad)
2011 Alfa Romeo 159 1.8 TBi 16V 146kW	7.7s	147kW	11.8 l/100km	6 l/100km	97%	*****
1998 Audi A4 2.8	7.1s	142kW	13.6 l/100km	7.0 l/100km	94%	***
2011 Audi A4 2.0T	6.9s	155kW	8.3 l/100km	5.3 l/100km	57%	*****
1998 BMW 328i	7.6s	142kW	12.5 l/100km	7.0 l/100km	79%	****
2011 BMW 325i	6.7s	160kW	9.8 l/100km	5.7 l/100km	42%	*****
1998 Mercedes C280 sport	7.8s	145kW	14.8 l/100km	7.4 l/100km	100%	****
2011 Mercedes C250 BlueEFFICIENCY (Auto)	7.2s	150kW	9 l/100km	5.6 l/100km	61%	*****

So, all the modern cars are safer. All the modern cars are faster (except Alfa, as fast). All the modern cars use substantially less fuel on the open road. All the modern cars use considerably less fuel in urban conditions. All the modern cars (except Alfa) has a much smaller increase when used in urban conditions than the old cars.

So, your uncle's 20 year old Opel can reach Durbs using only 25% more petrol, but try to commute from Fourways to Sandton every day...

AARTO, safety and red robot cameras

2011-05-03T09:03:00.000+02:00

So, you are approaching a robot that just turned orange, in an 80km/h zone, minding the speed limit. There is a permanent enforcement camera at the intersection.
The safe answer is to stop if possible, after also ensuring that you will not be rear-ended by the car behind.

Under AARTO with its demerit points, what should you do?
Well, charge code 1678 specifies that failing to stop behind the line carries a R500 fine and a demerit point. Charge code 1789 (doing 91-95km/h in an 80km/h zone) carries a R250 fine and no demerit points. Charge code 1790 (doing 96-100km/h in an 80km/h zone) carries a R500 fine and a demerit point.

Given that few cars can accelerate fast enough to reach 96km/h from 80km/h in a shorter distance than they can safely brake from 80km/h to 0km/h, it is clear that to minimize penalties under AARTO, the answer is to accelerate if there is even the slightest chance that you will be unable to stop behind the line.

Clearly AARTO is designed to keep us safe then! (Not).

And, just to make this clear, this is not a hypothetical situation. I have seen a traffic summons issued to a retired lady which just managed not to stop behind the stop line, but is clearly stationary and not endangering anybody, yet now has a demerit point. Had she gunned it instead, her driver's license would still be unencumbered.