So you are saying that you would rather have $5 than $5 and a slab of beer? The 'waste' of biofuel production is often worth as much as the fuel (just not as a liquid fuel for transportation). In most cases the remainder is useful as a fuel for electricity production or heat. In the future it may be possible to use this energy to provide the energy to produce the next crop, or simply to provide more energy. Discounting that simply because it isn't a goal product is misleading. It's like saying that gasoline has a higher energy value than ethanol, and neglecting the negatives such as the energy to run the refinery, the transportation costs, reliance on foreign oil and danger to the environment. To compare any fuels fairly we must include all the sides costs, and merging the total energy available seems a very valid claim to me.
As for corn farming for ethanol, I have never argued it is a good idea. Land that is suitable for corn is probably best used for that, and land that is not good enough for that can be used to grow a bio fuel such as wheat, sugarcase or hemp. I think arguing about corn ethanol is a strawman.
(And I think gasoline+otto is a stupid combination to begin with, so whether it is modified with ethanol is really moot)
Part of the problem is that we have had different exposure to biofuels and transport technologies. In Australia bioethanol and biodiesel are cost competitive with fossil oil, and when you can find an independant dealer, they often sell 80-20 ethanol considerably cheaper than regular unleaded. And diesel is a practical personal fuel here, as every petrol station has a diesel pump. (I should know, I share a DI ute with a friend)
perhaps you didn't read the article, but reading the intro and the conclusion it states that the payback is 1.34 (if people exagerate a claim from a link they give I have to conclude that they have an agenda). Anyway, that is corn ethanol, which is widely regarded as very inefficient. This article from NZ's dept of agriculture http://www.google.com/url?sa=t&ct=res&cd=2&url=htt p%3A//www.eeca.govt.nz/uploadedDocuments/WSL%2520b ioethanol%2520report%2520FINAL0.pdf&ei=XbMoQ5GKJsb c4QGXpfCZBw has corn at 1.24 (fairly close to the USDA figure), but puts lignocellulose ethanol at 5.3* return (and can use waste LC from paper pulp, food industries and paper kerbside recycling.
So I have to disagree (and yes, nobody will read this either:).
That's not really the right way of putting it. A peltier requires low voltage, high current DC, but really it is the efficiency of a device that dictates how much power is required - if it is 10% efficient and a compressor is 40% efficient, then you'll need 4 times as much power to achieve the same cooling effect.
However, compressor based approaches often have a huge starting current to turn the motor over and take a while to start cooling, whereas a peltier junction coolers start cooling immediately with a constant current draw (peltiers are also somewhat adjustable in output either through pwm or variable current methods). So although peltiers are far less efficient, they are also very useful for certain applications.
Now for a radiator pointing at the sky at night we radiate to a 4K sink. But the radiation is reflected and absorbed (which effectively reduces difference in temp - also known as insulation) by various gases, mainly water. So it must be a clear night, with low humidity. Deserts are ideal, and they generally have a swing of more than 60C each day as a result.
The air itself (O2 and N2) is practically transparent to the IR wavelengths and have no effect on the transmission, no matter what their temperature. The reason it doesn't work during the day is that the sun fills the sky with lots of IR scattered from dust and ice crystals, giving the sky a temperature of about 250K, which is nowhere near as effective for cooling things down.
It's an interesting phenomenon, and you can easily test it yourself - leave a foam box outside with some water in it on a clear cool night (adding some dye may help). Make sure that the water can't 'see' anything apart from the sky. for comparison put a similar body of water under a tree, or in a heat conductive container (plastic bucket).
A classic bush survival technique is to use a space blanket (the metal ones) to cover a black plastic sheet on the ground during the day to get it cold, then take it off at night - it will quite rapidly collect water - I collected 1L/m^2 when I tried it, which, over our 4m^2 tarp, was most of our daily water requirement.
Peltier junctions are very inefficient (10% of carnot, against 40% for a standard compressor type design). And none of that tech fits in the category of 'can be fixed locally'.
All of the technology would be doable with basic car repair technology, which is available pretty much everywhere. I can imagine the indians doing as a good a job with this approach as they do with the ubuiquitous petters: http://utterpower.com/petter.htm
Sometimes robust 19th century engineering can teach us something:)
Excellent, I was going to post this, but couldn't find any links. Thanks. You are right, this is a much simpler and more effective solution (sic) to the problem. Although ammonia is a little toxic, it is completely sealed in, and I expect that the probability of harm is far less than that from poorly kept meat.
If I were to use this design for a cheap system, I would probably leave out the tap though - cold can equally well be stored as ice, and the tap provides a weak point in the system.
It works because radiation transfer goes as the 4th power of the different between the two objects. And space is very cold. I've found ice in polystyrene foam boxes here in melbourne when the minimum temp was 6C - as long as there is nothiner between you and space it can get very cold indeed!
You're right: WTF! Microwaves or soundwaves? If they can control the weather at this energy, why aren't they holding the world ransom for say, 1 million dollars!, or something?
Rotting vegetation and termites are carbon neutral and are irrelevant in this discussion. Volcanos produce (as quoted earlier) about 1% of the CO2 of human activity. Remember that we are talking about new carbon in the atmosphere, not stuff that is going around and around. (And ikea don't use indonesian clearfelled hardwood either - they use scandinavian plantation beech and softwoods)
Inventing facts to support your world view is just silly. (Mind you, so is responding to ACs)
That's all very nice, but your starting value is the total amount of rubbish produced. In my experience the plastic portion of the waste stream is quite small in comparison to that weight of glass, steel, compostable and aluminium components; which are better recycled directly rather than through an energy intensive process like TD. The plastics and non-compostable mass of my waste stream is about 5% of the total, bringing your estimate down to 200000 barrels per year - barely a sniff.
You claim that the energy to remove the water is insignificant. The amount of energy required to vapourise water is equivalent to heating the water to 600C, so it is exactly the same amount of energy as my original estimate (namely 2.5kJ/g). Unless TD can recover the majority of the thermal energy(which it can't in my understanding), this process is akin to direct thermal desalination of seawater in terms of total energy use, i.e. lots.
We know what the energy value of human waste is - it is burnt in some countries as a fuel. It is about 40% the calorific value of wood (about 8MJ/kg). I estimate I produce about 1kg of human waste per day, or about 2.2kWhr of energy per day. Assuming perfect conversion (the 8MJ/kg is dry value, incidently), I could just about run my computer all day off my personal waste output. I certainly couldn't drive to work.
The big problem people have with estimating energy returns on these kinds of energy sources is that they always want to use the big numbers by combining over a city. If you can't sustain yourself realistically with a given energy source, scaling it up doesn't help. (Did you hear about the shop that made a loss on every thing they sold? They made it up in volume). Yes, a 10MW plant is a nice thing, but if you can only make one of these per city you would be better off examining a different source.
Industrial waste is better recycled directly in most cases (more efficiently) than by converting it into oil. Stop believing everything of a technology that as yet has only demonstrated something akin to rendering glue. Yes, it will be a useful ally in the global reduction of energy, no, it will not be a useful part of the future energy mix.
Another way to look at it is that to produce each item of plastic, glass, food or steel requires about 10 times as much energy as the embodied energy of the item itself. It is quite clear that even a perfectly efficient extraction process is not sustainable.
Now, if we were talking about producing oils for use as feed stocks in a future world without fossil oil - I'd be the first person to promote TD. For energy, no go.
I'm sure it's possible to make oil out of sewage. But it's not going to have much energy left in it - already the body has pulled most of the calorific value out. For a start you need to get all the water out - I bet that the energy required to dry it is probably more than the embodied energy (and if you dry it with the sun, well you'll get more energy out of using that sunlight directly).
The garbage argument was covered in my original post, I thought. Basically we can put a bound on the available energy by considering the relative quantities of crude required to provide energy and to provide plastics feedstock.
TD will be more important as a way to avoid landfill and to capture valuable elements than as a source of energy.
You are wrong. The payback for bio-ethanol is about 2.5*. The payback for biodiesel is about 3.5*. Biodiesel rises to nearly 5 in suitable climates when the straw is also burnt for energy.
You have to put energy into to get access to any energy (this is called activation energy). It is the ration between the input and the put that is important, and for crude (Acording to posters yesterday) this is about 5*.
Actually, LEDs aren't very efficient, last I looked they were more efficient than incandesent (but that's not hard:), but less than compact fluoros. The high pressure sodium lights, which have that characteristic yellow glow are probably the most efficient light source out there. Particularly as they emit solely in one of the more sensitive regions of our vision. This means that the effective illumination per watt is even greater. There is a cost to having white light.
LEDs make lots of sense for specialist task lighting - I have leds set up to illuminate things at night which means I made our whole house navigable in the darkest night using only 1W of electricity. Far better than lots of electroluminescent night lights, and not as sickly:)
The oil from TD is clearly less energetic than its inputs. And we use far more oil as fuel than for plastics and whatnot (I think products are about 7% of total oil consumption). So if we used our current 50 year supply of petro-waste to produce oil we'd use it all up in less than 3 years. We could use TD to make oil out of say biomass, but then we'd be better off with biodiesel, which is a more efficient conversion.
That would be true if we were to use genetically modified alga. However, there is already some well understood species we can use that deliver about 50% of their body mass in oils. So personally I would like to steer away from GM tech if we could already do a good job with what we've got.
(Why is it that people need to introduce new technologies to solve problems that are already solved with old and simple technologies?)
I agree 100%. I was merely pointing out that if fusion energy is plentiful then there are many ways to produce chemical stored energy from electricity.
Growing under lights is useful to study in any case, as we'll be doing that when we first build our moonbase or marsbase. 600L per acre is rather poor considering the energy input. I estimate around 10TJ of energy coming in, so we're looking at something that is less than a percent efficient.
The algae idea has been around a while. Arizona has some sewage treatment plants based on this approach already - a friend of mine was involved in the biological design and bounces regularly about this approach. A far better place to grow the algae is on the surface of the ocean, where the water is plentiful and has exactly the right chemistry for the alga.
There are fairly simple chemical processes for converting H2 and CO2 into oil. They just aren't economic because we produce H2 from oil at the moment. If H2 were cheap, producing oil replacements would be relatively straighforward.
Another approach is to electrolyse the CO2 into carbon and oxygen, then react this with water to produce oil. However, that technology was developed for producing oil from coal, and there is plenty of coal around, so unless fusion power is surprisingly cheap, we'll probably just use coal.
Converting methane into propane and butane is already done on a large scale, and in some countries these gases are already commonly used as car fuels (LPG in australia).
Finally, if fusion electricity is cheap enough, we can simply grow very dense crops under electric lighting and convert the resulting bio-oils to biodiesel.
Just out of curiosity, what wonderful UI advances have you seen for Windows that would put Looking Glass in its place? Personally I think Looking Glass is overhyped shit, but I'm interested in what you think is better, as you clearly know of some fancy new tech and I'm missing out.
To me far better examples of UI improvements are things like google search/beagle/spotlight, workspaces and command lines...
Not sure of the accuracy of more recent chips, but the original gave 5 bits accuracy, which can be polished in 4 loops of netwon's method.
I'm not sure what uses there are for highspeed scalar sqrts these days, with graphics being handled by the GPU and audio being handled by altivec with its own sqrt estimate and similar performance to the cell: http://developer.apple.com/hardware/ve/algorithms. html
Notching is a wear pattern whereby certain ratios common in driving wear a groove in the CVT cones, making the car tend to lose power in that band. I imagine it takes 5 years or more to happen.
teach me to preview my comments - all my nice virtual html tags were removed...
"So, what do you do?" (excited)"I'm a mathematician working on optimisation techniques" "Really? What does that mean?" (insert standard general audience research abstract here)
"So, what do you do?" (staring at shoes)"I'm a mathematician working on optimisation techniques" "Oh, I was never any good at math in school." (change topic quickly)
Interesting. When I tell people that I'm a mathematician they are usually impressed. I get the feeling that they think I'm some kind of Jeff Goldblume or something:) Perhaps it is the way you say it? I think people watch more for how you say it that what you say.
"So, what do you do?" "I'm a mathematician working on optimisation techniques" "Really? What does that mean?"
"So, what do you do?" "I'm a mathematician working on optimisation techniques" "Oh, I was never any good at math in school."
Slashdot Mirror
Agreed.
We all make mistakes (me more than most :).
So you are saying that you would rather have $5 than $5 and a slab of beer? The 'waste' of biofuel production is often worth as much as the fuel (just not as a liquid fuel for transportation). In most cases the remainder is useful as a fuel for electricity production or heat. In the future it may be possible to use this energy to provide the energy to produce the next crop, or simply to provide more energy. Discounting that simply because it isn't a goal product is misleading. It's like saying that gasoline has a higher energy value than ethanol, and neglecting the negatives such as the energy to run the refinery, the transportation costs, reliance on foreign oil and danger to the environment. To compare any fuels fairly we must include all the sides costs, and merging the total energy available seems a very valid claim to me.
As for corn farming for ethanol, I have never argued it is a good idea. Land that is suitable for corn is probably best used for that, and land that is not good enough for that can be used to grow a bio fuel such as wheat, sugarcase or hemp. I think arguing about corn ethanol is a strawman.
(And I think gasoline+otto is a stupid combination to begin with, so whether it is modified with ethanol is really moot)
Part of the problem is that we have had different exposure to biofuels and transport technologies. In Australia bioethanol and biodiesel are cost competitive with fossil oil, and when you can find an independant dealer, they often sell 80-20 ethanol considerably cheaper than regular unleaded. And diesel is a practical personal fuel here, as every petrol station has a diesel pump. (I should know, I share a DI ute with a friend)
perhaps you didn't read the article, but reading the intro and the conclusion it states that the payback is 1.34 (if people exagerate a claim from a link they give I have to conclude that they have an agenda). Anyway, that is corn ethanol, which is widely regarded as very inefficient. This article from NZ's dept of agriculturet p%3A//www.eeca.govt.nz/uploadedDocuments/WSL%2520b ioethanol%2520report%2520FINAL0.pdf&ei=XbMoQ5GKJsb c4QGXpfCZBw
:).
http://www.google.com/url?sa=t&ct=res&cd=2&url=ht
has corn at 1.24 (fairly close to the USDA figure), but puts lignocellulose ethanol at 5.3* return (and can use waste LC from paper pulp, food industries and paper kerbside recycling.
So I have to disagree (and yes, nobody will read this either
That's not really the right way of putting it. A peltier requires low voltage, high current DC, but really it is the efficiency of a device that dictates how much power is required - if it is 10% efficient and a compressor is 40% efficient, then you'll need 4 times as much power to achieve the same cooling effect.
However, compressor based approaches often have a huge starting current to turn the motor over and take a while to start cooling, whereas a peltier junction coolers start cooling immediately with a constant current draw (peltiers are also somewhat adjustable in output either through pwm or variable current methods). So although peltiers are far less efficient, they are also very useful for certain applications.
I actually had a discussion about this some years ago here on slashdot :)
/ stefan.html
The energy transmitted by a blackbody is proportional to the difference of the fourth power of the temperature of the two bodies: http://hyperphysics.phy-astr.gsu.edu/hbase/thermo
Now for a radiator pointing at the sky at night we radiate to a 4K sink. But the radiation is reflected and absorbed (which effectively reduces difference in temp - also known as insulation) by various gases, mainly water. So it must be a clear night, with low humidity. Deserts are ideal, and they generally have a swing of more than 60C each day as a result.
The air itself (O2 and N2) is practically transparent to the IR wavelengths and have no effect on the transmission, no matter what their temperature. The reason it doesn't work during the day is that the sun fills the sky with lots of IR scattered from dust and ice crystals, giving the sky a temperature of about 250K, which is nowhere near as effective for cooling things down.
It's an interesting phenomenon, and you can easily test it yourself - leave a foam box outside with some water in it on a clear cool night (adding some dye may help). Make sure that the water can't 'see' anything apart from the sky. for comparison put a similar body of water under a tree, or in a heat conductive container (plastic bucket).
A classic bush survival technique is to use a space blanket (the metal ones) to cover a black plastic sheet on the ground during the day to get it cold, then take it off at night - it will quite rapidly collect water - I collected 1L/m^2 when I tried it, which, over our 4m^2 tarp, was most of our daily water requirement.
Peltier junctions are very inefficient (10% of carnot, against 40% for a standard compressor type design). And none of that tech fits in the category of 'can be fixed locally'.
t ml
:)
I think the crosley icyball is a far better solution though: http://www.ggw.org/~cac/IcyBall/crosley_icyball.h
All of the technology would be doable with basic car repair technology, which is available pretty much everywhere. I can imagine the indians doing as a good a job with this approach as they do with the ubuiquitous petters: http://utterpower.com/petter.htm
Sometimes robust 19th century engineering can teach us something
Excellent, I was going to post this, but couldn't find any links. Thanks. You are right, this is a much simpler and more effective solution (sic) to the problem. Although ammonia is a little toxic, it is completely sealed in, and I expect that the probability of harm is far less than that from poorly kept meat.
If I were to use this design for a cheap system, I would probably leave out the tap though - cold can equally well be stored as ice, and the tap provides a weak point in the system.
It works because radiation transfer goes as the 4th power of the different between the two objects. And space is very cold. I've found ice in polystyrene foam boxes here in melbourne when the minimum temp was 6C - as long as there is nothiner between you and space it can get very cold indeed!
You're right: WTF! Microwaves or soundwaves? If they can control the weather at this energy, why aren't they holding the world ransom for say, 1 million dollars!, or something?
Rotting vegetation and termites are carbon neutral and are irrelevant in this discussion. Volcanos produce (as quoted earlier) about 1% of the CO2 of human activity. Remember that we are talking about new carbon in the atmosphere, not stuff that is going around and around. (And ikea don't use indonesian clearfelled hardwood either - they use scandinavian plantation beech and softwoods)
Inventing facts to support your world view is just silly. (Mind you, so is responding to ACs)
That's all very nice, but your starting value is the total amount of rubbish produced. In my experience the plastic portion of the waste stream is quite small in comparison to that weight of glass, steel, compostable and aluminium components; which are better recycled directly rather than through an energy intensive process like TD. The plastics and non-compostable mass of my waste stream is about 5% of the total, bringing your estimate down to 200000 barrels per year - barely a sniff.
You claim that the energy to remove the water is insignificant. The amount of energy required to vapourise water is equivalent to heating the water to 600C, so it is exactly the same amount of energy as my original estimate (namely 2.5kJ/g). Unless TD can recover the majority of the thermal energy(which it can't in my understanding), this process is akin to direct thermal desalination of seawater in terms of total energy use, i.e. lots.
We know what the energy value of human waste is - it is burnt in some countries as a fuel. It is about 40% the calorific value of wood (about 8MJ/kg). I estimate I produce about 1kg of human waste per day, or about 2.2kWhr of energy per day. Assuming perfect conversion (the 8MJ/kg is dry value, incidently), I could just about run my computer all day off my personal waste output. I certainly couldn't drive to work.
The big problem people have with estimating energy returns on these kinds of energy sources is that they always want to use the big numbers by combining over a city. If you can't sustain yourself realistically with a given energy source, scaling it up doesn't help. (Did you hear about the shop that made a loss on every thing they sold? They made it up in volume). Yes, a 10MW plant is a nice thing, but if you can only make one of these per city you would be better off examining a different source.
Industrial waste is better recycled directly in most cases (more efficiently) than by converting it into oil. Stop believing everything of a technology that as yet has only demonstrated something akin to rendering glue. Yes, it will be a useful ally in the global reduction of energy, no, it will not be a useful part of the future energy mix.
Another way to look at it is that to produce each item of plastic, glass, food or steel requires about 10 times as much energy as the embodied energy of the item itself. It is quite clear that even a perfectly efficient extraction process is not sustainable.
Now, if we were talking about producing oils for use as feed stocks in a future world without fossil oil - I'd be the first person to promote TD. For energy, no go.
I'm sure it's possible to make oil out of sewage. But it's not going to have much energy left in it - already the body has pulled most of the calorific value out. For a start you need to get all the water out - I bet that the energy required to dry it is probably more than the embodied energy (and if you dry it with the sun, well you'll get more energy out of using that sunlight directly).
The garbage argument was covered in my original post, I thought. Basically we can put a bound on the available energy by considering the relative quantities of crude required to provide energy and to provide plastics feedstock.
TD will be more important as a way to avoid landfill and to capture valuable elements than as a source of energy.
You are wrong. The payback for bio-ethanol is about 2.5*. The payback for biodiesel is about 3.5*. Biodiesel rises to nearly 5 in suitable climates when the straw is also burnt for energy.
You have to put energy into to get access to any energy (this is called activation energy). It is the ration between the input and the put that is important, and for crude (Acording to posters yesterday) this is about 5*.
The payback for wind is about 20*.
Actually, LEDs aren't very efficient, last I looked they were more efficient than incandesent (but that's not hard :), but less than compact fluoros. The high pressure sodium lights, which have that characteristic yellow glow are probably the most efficient light source out there. Particularly as they emit solely in one of the more sensitive regions of our vision. This means that the effective illumination per watt is even greater. There is a cost to having white light.
:)
LEDs make lots of sense for specialist task lighting - I have leds set up to illuminate things at night which means I made our whole house navigable in the darkest night using only 1W of electricity. Far better than lots of electroluminescent night lights, and not as sickly
The oil from TD is clearly less energetic than its inputs. And we use far more oil as fuel than for plastics and whatnot (I think products are about 7% of total oil consumption). So if we used our current 50 year supply of petro-waste to produce oil we'd use it all up in less than 3 years. We could use TD to make oil out of say biomass, but then we'd be better off with biodiesel, which is a more efficient conversion.
That would be true if we were to use genetically modified alga. However, there is already some well understood species we can use that deliver about 50% of their body mass in oils. So personally I would like to steer away from GM tech if we could already do a good job with what we've got.
(Why is it that people need to introduce new technologies to solve problems that are already solved with old and simple technologies?)
I agree 100%. I was merely pointing out that if fusion energy is plentiful then there are many ways to produce chemical stored energy from electricity.
Growing under lights is useful to study in any case, as we'll be doing that when we first build our moonbase or marsbase. 600L per acre is rather poor considering the energy input. I estimate around 10TJ of energy coming in, so we're looking at something that is less than a percent efficient.
The algae idea has been around a while. Arizona has some sewage treatment plants based on this approach already - a friend of mine was involved in the biological design and bounces regularly about this approach. A far better place to grow the algae is on the surface of the ocean, where the water is plentiful and has exactly the right chemistry for the alga.
There are fairly simple chemical processes for converting H2 and CO2 into oil. They just aren't economic because we produce H2 from oil at the moment. If H2 were cheap, producing oil replacements would be relatively straighforward.
Another approach is to electrolyse the CO2 into carbon and oxygen, then react this with water to produce oil. However, that technology was developed for producing oil from coal, and there is plenty of coal around, so unless fusion power is surprisingly cheap, we'll probably just use coal.
Converting methane into propane and butane is already done on a large scale, and in some countries these gases are already commonly used as car fuels (LPG in australia).
Finally, if fusion electricity is cheap enough, we can simply grow very dense crops under electric lighting and convert the resulting bio-oils to biodiesel.
"Where else can you get a great desktop environment that just works, along with a built-in Unix-like OS?"
Here: Ubuntu Linux
Muon catalysed fusion.
Just out of curiosity, what wonderful UI advances have you seen for Windows that would put Looking Glass in its place? Personally I think Looking Glass is overhyped shit, but I'm interested in what you think is better, as you clearly know of some fancy new tech and I'm missing out.
To me far better examples of UI improvements are things like google search/beagle/spotlight, workspaces and command lines...
PowerPC (and presumably POWER) have had frsqrte and fres for ages: http://publibn.boulder.ibm.com/doc_link/en_US/a_do c_lib/aixassem/alangref/frsqrte.htm
. html
Not sure of the accuracy of more recent chips, but the original gave 5 bits accuracy, which can be polished in 4 loops of netwon's method.
I'm not sure what uses there are for highspeed scalar sqrts these days, with graphics being handled by the GPU and audio being handled by altivec with its own sqrt estimate and similar performance to the cell:
http://developer.apple.com/hardware/ve/algorithms
Notching is a wear pattern whereby certain ratios common in driving wear a groove in the CVT cones, making the car tend to lose power in that band. I imagine it takes 5 years or more to happen.
teach me to preview my comments - all my nice virtual html tags were removed...
"So, what do you do?"
(excited)"I'm a mathematician working on optimisation techniques"
"Really? What does that mean?"
(insert standard general audience research abstract here)
"So, what do you do?"
(staring at shoes)"I'm a mathematician working on optimisation techniques"
"Oh, I was never any good at math in school."
(change topic quickly)
Interesting. When I tell people that I'm a mathematician they are usually impressed. I get the feeling that they think I'm some kind of Jeff Goldblume or something :) Perhaps it is the way you say it? I think people watch more for how you say it that what you say.
"So, what do you do?"
"I'm a mathematician working on optimisation techniques"
"Really? What does that mean?"
"So, what do you do?"
"I'm a mathematician working on optimisation techniques"
"Oh, I was never any good at math in school."