Wednesday May 16, 2012 | ${log.root}/lowem.log Inflation, Investing and Everything |
|
This is an initial speed estimate of the Ferrari that rammed into the taxi at Bugis Junction, Singapore early in the morning of 5th May 2012. The story and video can be viewed here. Other links, opinions and commentary can be found online on Facebook and other social networking sites. The usual disclaimers apply : I am not an expert in this area. I am filing this under the Tech section while utilizing appropriate technical tools available to try to come to an estimate, and to show how this could be derived based on currently available data. Speed = distance divided by time, so we need to estimate both the distance travelled and the time taken. Let's start with the time factor. We shall first look at the video recorded by the other taxi that fortunately escaped the collision. The first frame that shows the Ferrari crossing the second line of the pedestrian crossing is as follows :
7 frames later, the Ferrari has crashed into the side of the taxi :
If we examine the meta-data on the video, we note that it is running at 25 frames per second (25fps) :
7 frames divided by 25 frames per second gives us 0.28 seconds. We need to divide further by 60 to get it in minutes, and divide again by 60 to normalize it into the hours unit which is 7.778E-5 hr, which we then note down. Next, we need to estimate the distance travelled. Whereas a more accurate method would be of course to go down to the scene and carry out measurements on the spot, the next best estimate we might be able to get is from using the measurement tool on Google Maps while setting it to the maximum possible zoom level without going into Street View mode. This is as follows :
We have 14 meters, give or take perhaps a meter or two. Normalizing it into the kilometer unit, we get 0.014km. Hence, the initial estimate of the speed is 0.014 km / 7.778E-5 hour = 180 km/h. No doubt there could be questions about counting the frames in the video. Each frame is 0.04 seconds long, and let's say the count is off by 1 frame or less (i.e. between 6 and 7 frames). In addition, the line drawn with the measurement tool on Google Maps could be perhaps shorter by a meter or two (give or take). We plug these figures into a spreadsheet and arrive at a matrix of various speed estimates :
As can be seen, the speed estimates range from 154 to 210 km/h, depending on the two variables of frame count and distance travelled. The speed of the Ferrari based on the above estimation is thus 180 km/h +/- 30 km/h. (2012-05-16 21:48:21 SGT)
[Tech]
Permalink
Comments [1]
Spread the word. You can write in too with your own variation. To UOB Singapore via internal email (under Email Us icon after login to internet banking) : "As Earth Hour is today and Earth Day is coming up not too long after this, I am writing in to request that UOB Singapore consider implementing electronic statements for all UOB accounts, whether credit cards or savings accounts, in order to save on paper and transport costs as well as the associated waste and greenhouse emissions involved. Of all the banks I am a customer of, including DBS/POSB, Citibank, OCBC etc, UOB is the only bank that has not provided such a facility as far as I know. Hence, please do consider providing this capability or if there is one already, please point me to where I can sign up for it. Thank you." (2012-03-31 12:11:32 SGT)
[Env]
Permalink
So I've been writing in to a Minister once again. Probably not the first or last person to bring up these simple facts to Tharman Shanmugaratnam, Minister for Finance of Singapore, but I thought I would contribute my two cents (ha, ha). Here it goes : ... Dear Minister for Finance, I am writing in my personal capacity regarding the recent DBS/POSB ATM fraud case. As we have seen, the two-factor authentication mechanism of ATM/NETS cards has been defeated. The perpetrators have managed to successfully obtain both factors of authentication : something you have (ATM card details via the card skimming device), and something you know (ATM PIN via a strategically-located pinhole camera). It has been proven time and again that the magnetic stripe data such as those being used in the ATM and NETS cards in Singapore is quite easily copied. This magnetic stripe technology was invented over 50 years ago and is able to contain at most a few bytes of information, and has no processing capability of any sort, nor any form of cryptographic technology to resist cloning or tampering. In short, this is a completely outdated technolgy that the whole of our nation is using for our daily financial transactions, at thousands of retail establishments, ATM's, top-up kiosks, and so on. Something needs to be done about this glaring security vulnerability. As you can see from the link below, Malaysia's Maybank is already ahead of Singapore in this respect : The Maybank ATM card includes an embedded smartcard chip, which as their website states, offers "Increased security, as the smart chip is tamper-resistant and the data stored is harder to extract and copy". This is the key point : smartcard technology, well established by now, mitigates many of the shortcomings of magnetic stripes. It is much harder to simply clone compared to magnetic stripes and requires sophisticated and intrusive physical attacks to get to the private key data contained in the smartcard chip. The technology currently required to carry out such attacks does not fit in a simple card skimmer that can be installed in an ATM card slot. Other security measures such as revocation of the certificates of any particular set of smartcards can be carried out on demand. In addition, manufacturers have already included a number of on-board security measures in smartcard processor chips to resist cloning and tampering. Hence, in the interest of increased security for banking customers, I would like to suggest that MAS work towards regulations requiring all banks issuing ATM and NETS cards to include embedded smartcard technology such as the EMV technology being used by VISA and other credit card issuers. As an additional security measure, I would also like to suggest industry-wide regulation to eventually disallow transaction fallback to magnetic strip reading, thus removing one of the main factors contributing to ATM card fraud. The ultimate goal is for ATM/NETS cards to be issued *without* any form of magnetic stripes whatsoever, and ATM's as well as POS terminals to be updated to smartcard-acceptance only, thus finally retiring this outdated 50-year old magnetic stripe technology. Thank you for your attention. Low Ee Mien (Mr) - Sure, smartcards are not impossible to defeat, but they are much harder than the increasingly laughable magnetic stripes, with card skimmers being more readily available nowadays, even apparently custom fitted to closely resemble the "anti-intrusion" devices installed over the ATM card slots. Some have said that 3D printers have now been used to create the exact shapes of these "anti-skimming" devices, or Fraudulent Device Inhibitor (FDI), in an ironic application of cutting-edge technology being used to defeat a simple physical device that was meant to guard against just such occurrences of card skimmers being installed. The measures announced by the bank are well and fine, such as blocking foreign ATM withdrawals for folks who have not used them overseas, sending out SMS messages and so on, but all these reek of reactionary and stop-gap measures. I don't quite agree with the DBS CEO's claim that replacing the magstripes with smartcard chips has its own set of problems, such as : "The problem with that is it's a huge inconvenience to customers. When you go to the US, they don't accept chip cards". My man, the world is moving away from this outdated magstripe technology, I am frankly surprised and amazed simply that Singapore has not gone along and our neighbour Malaysia has. Fact is, we are actually behind Maybank Malaysia in this area and there's quite a bit of catching up to do. (2012-01-16 11:41:39 SGT)
[Musings]
Permalink
Hello World Android! Took me long enough to get going, but here's my first honest-to-goodness Android app, running in the AVD emulator : And here it is running on my Samsung Galaxy S2 : Setting up the development environment was simple enough, follow Google's instructions, get the latest Eclipse (yes, I admit I've been a little out of touch lately), get the Android SDK, download the necessary files, libraries and the Android emulator and off I went. The emulator did take all of 5 minutes or perhaps a bit more than that to boot up on my Acer laptop which up until now had yet to show its age, being perfectly useful for browsing and Microsoft Office and even the occasional VMWare. But now a 2.x GHz 8GB RAM replacement number looks mighty enticing. Running it on the actual device was also simple enough : look for the hello-android.apk in the bin folder, Gmail it to myself, install (application side-loading on the Galaxy S2 was enabled *out of the box*), take screenshot, Gmail screenshot back to myself, and voila. So goes the saying : "You have taken your first step into a larger world." Or something along these lines. (2011-12-14 00:27:58 SGT)
[Java]
Permalink
channelnewsasia.com, foxbusiness.com : Shell's fire-hit Pulau Bukom oil refinery may remain shut for at least a month. Shell is tight-lipped about how much money is going up in smoke due to the incident at its biggest refinery. But some analysts said disrupted production would affect regional supplies of Shell's products in the short term. Shell's declaration on Sun 2 Oct 2011 of force majeure on the supply of oil products to some customers following a fire at its 500,000-barrel-a-day Singapore refinery has done little to provide answers on how long refining operations will remain disrupted. While petroleum markets have already priced in a short-term supply disruption, with gasoline, middle distillates and fuel oil all up $2-3 per barrel against crude oil prices since the fire, the declaration has exacerbated supply worries. Shell has not provided a forecast on when repairs to the refinery - the company's largest - will be complete, although it has said refining units at the Pulau Bukom complex were not damaged by the blaze which started in a pumping station near storage tanks and burned from Wednesday to Friday. The fire may have already dented demand for Middle East crude oil. A Singapore trader said Shell may have cancelled 4 million barrels of Oct 2011 crude from Saudi Arabia. It remains difficult to assess the full impact of Shell's supply disruption. Shell said it has started an investigation over the cause of the fire, and won't restart the refinery until safety is ensured, despite damage being limited to a small area. - Though this has been one of the biggest stories out of Singapore for the past week or so, there are still many questions that are as yet unanswered, the exact cause of the fire only being one of them. The other issues include the possibility of tightness downstream through the supply chain, in particular gasoline (or petrol as we call it here in Singapore), as well as diesel. Another quite valid question is of course something that people might ask me, such as "should I go and top off my fuel tank right away, reduce my driving and in general prepare for an all-out calamity?" Let's try to estimate this using best guesses. The Shell Bukom refinery operates at 500,000 barrels per day, which is 500kbpd, or 0.5mbpd as the industry as well as we peakoilers usually refer to it. Reportedly, 90% of that is exported, which means that 10% or 50kbpd is for local consumption. Assuming that half of it is utilized for gasoline and diesel production, that works out to 25 kbpd worth, or close to 4 million liters per day. Assuming that each vehicle has a fuel tank of 50 liters, we come to an impact estimate on 79,500 vehicles per day. The total vehicle population in Singapore is 952,221 vehicles (as at Jul 2011 according to the Singapore Department of Statistics), hence the impact will be felt on about 8.3% of the Singapore vehicle population, which roughly speaking, would be 1 out of every 12 vehicles. As you can see, this is not an insignificant impact though it shouldn't be calamitous. Given the storage buffers throughout the supply chain, we might not expect to see an immediate impact right away. From past observations of petrol price movements, we might expect changes if any to hit within 1 to 2 weeks' time. This isn't much of a forecast, but we might expect petrol and diesel prices to move upward in this timeframe. As they say, pricing happens at the margins, and this is quite a margin. Of course, many things might happen in a dynamic market. For all we know, it might all come down to who's bidding and who's winning. Singapore may export less distillates as a whole, with price increases passed on to local motorists, and the rest of the impact spread out among the wider Asian region. The extreme opposite could also be true, if the locals balk at increased prices, and so on. At the point of writing, nobody knows for sure. We should have a much clearer picture soon, within 2 weeks at most. (2011-10-03 21:31:47 SGT)
[Energy]
Permalink
Switzerland opened a new round in the global currency war on 6 Sep 2011 as the Swiss National Bank's decision to cap the Swiss franc for the first time since 1978 marked a bid to protect trade hurt by record currency strength against the euro and dollar. The Swiss central bank said it is "prepared to buy foreign currency in unlimited quantities" to keep the euro above 1.20 francs. The franc plunged a record 8.1% against the euro on the SNB's unilateral move, putting it head-to-head with the $4 trillion-a-day forex market that drove the franc up more than 16% against 9 major peers in the past year. The move may help stabilize markets by forcing investors to return to riskier assets, said Jim O'Neill, chairman of Goldman Sachs in London. The initiative may leave Norway and Sweden vulnerable to gains in their currencies as countries such as Brazil and Japan fight to limit appreciation amid a flight from the euro debt crisis and near-zero US interest rates. Led by China, all of Asia's 10 biggest economies last year sought to influence their own exchange rates to aid exporters as the dollar fell. HSBC recommended Norway's currency as an alternative after the SNB's action. The krone has strengthened 4.5% against its 9 major peers over the past year. - This was one big announcement by the Swiss. At one stroke, EUR/CHF jumped from approximately $1.10 to $1.20 and stayed there. That's a gap up of around 10 cents, or in forex trading terms, 1000 pips. In the world of leveraged foreign currency trading, in large enough amounts, profits can be made trading in as little as 1-2 pips, while very good money can be made (or lost) in 10-20 pips. Hence 1000 pips is analogous to a Richter-scale 10.0 earthquake. So the Swiss have gone nuclear on their own currency. While the exact timing of the announcement might have been a surprise, there had been hints along the way in recent weeks as the Swiss authorities had been openly discussing possible pegs to the euro and such. Investors may wish to take note that when the Japanese kicked off the current phase of the currency wars back in Sep 2010, gold prices had gone up nearly 60% from the $1200 region to over $1900 in the past year. This next round in the currency wars could eventually take prices of gold, commodities and other related assets into literally unchartered territory. Do not expect immediate jumps across asset classes though - we shall see how this plays out over time. See also : 1. Gold price as a damped spring (2011-09-08 00:44:32 SGT)
[Biz]
Permalink
Most popular blog postings on lowem.log : 1. Singapore MRT rail network length to double by 2020 Featured articles on lowem.log : 1. Book review : Shut Down by William Flynn |
|
||||||||||||||||||||||||||||||