Sunday, 14 August 2016

Robotic storage

As with many industries, the future of self storage is robotics. Here's how it might look.

It's the end of the summer, you've just returned from a camping trip and are not planning to go camping again till next spring. You tap a couple of buttons on your smartphone an a couple of hours later a delivery drone arrives at your door with a couple of empty plastic storage crate. You load your camping equipment into the boxes, drop the boxes onto the drone, and forget about it till next spring. Whilst you're packing for your next camping trip, another couple of taps on your phone and your camping equipment is back with you.

You never knew where it was stored, and you didn't much care. All that you knew was that it was cheap, hassle free, and never more than an hour away.

To summarize the features of this technology:
  • Storage is containerized (not ISO shipping containers though!)
  • Boxes are collected from and delivered to customers' homes (or other location of choice)
This solution is actually possible with existing technology: we already have self-driving cars, and robots that can manipulate stand-sized packages (the storage crate).

Other than the customer-service benefit, there is a significant benefit to the storage company in that the storage does not need to be located as near to the customer as current self-storage solutions. This potentially allows cheaper land to be used, and larger buildings, resulting in economies of scale.

With more advanced robots

With more advanced robots it should be possible to handle non-containerized objects provided they are able to fit into the transportation drone. In the nearer term, the customer may load these onto the delivery drone at the customer's home and specialist drones will unpack. However it is likely to become possible that the delivery drone will arrive with a packing bot onboard. They packing bot would enter the customer's home to pick up the items, and could potentially wrap and box smaller items.

Consequences of such technology

With this service we can more easily own more things than we can fit into our homes, and as such it is likely that our propensity to consume could increase. However, delivery drones can also make it easier to rent items, which would tend to counter this effect.

Penalty for acquisitive crime

In addition to any appropriate incarceration or community service I think the financial penalty for acquisitive crime should be the confiscation of the offender's assets such as the person has no more than median net assets of the population.

Advantages:

  • The recipients of such penalty cannot argue that they are impoverished
  • There is no difficulty in proving that assets are the proceeds of crime

Anthro-pessimism

I consider myself a techno-optimism, albeit of a slightly more reserved kind that the most bullish who fit that description. Whilst my techno-optimism is tempered to some degree by inherent attributes of technology which make them dangerous, it is tempered by a much greater degree by what I have begun to call anthro-pessimism.

It is not a new idea to attribute the negative aspects of technology to man's use of them, but I think it is important to extract the issue of anthro-pessimism from the techno-optimism / techno-pessimism debate, not least because if we can define and characterize anthro-pessimism, we may be able to solve it.

And when we do, we can stop worrying quite so much about technology. In fact it's likely that technology will be able to help us solve the causes of anthro-pessimism.

So how might we define anthro-pessimism? Hollywood (and George W. Bush) would have us call it evil. Others might blame it on selfishness.

Selfishness is not a bad place to start, but in fact people in democracies sometimes vote against things that are not in their self interest (out of ignorance or irrationality, not altruism). Hence I've settled on the following factors as the root cause of my anthro-pessimism:

  • selfishness
  • lack of education / ignorance
  • cognitive biases / rationality
Selfishness - to a small extent this factor has been addressed by the structure of free market / meritocratic societies. It may also be an inherent characteristic that is hard to eliminate. However doing so may not be necessary if the other two factors are adequately addressed. If not it may also be mitigated to some degree by optimizing the structure of society.

Lack of education / ignorance - this is a huge area and in terms of human history is only very recently being addressed. One of the ways current education fails the most is in giving students a broad understanding of the world around them, how it got to be this way and how it may develop in the future; the teaching of history is often particularly narrow and biased.

Cognitive biases / rationality - as I included ignorance as one of my causal factors, perhaps we'd be OK it everyone were educated to the level of a university professor? No so unfortunately: professors are not immune to the cognitive biases that plague human cognition, and are hence not rational, and will not necessarily make optimal decisions.

The three factors can be restated in less human terms as follows:
  • Motivation for decision
  • Access to information
  • Tools to process the information to make a decision that matches the motivation
There is another rarer cause for anthro-pessimism that isn't covered by the three factors above mentioned: a mental health condition that would motivate an actor (or actors) such that despite an absence of ignorance and irrationality, and even considering the consequences of action that would be imposed upon the actor (or actors) by society for committing a harmful act, would still be sufficiently motivated to do so. I am not entirely sure whether such a mental condition exists as I strongly suspect most historical cases that may appear similar would actually so significant ignorance or irrationality, but it is worth including for completeness. It also potentially illustrates a blurry line between selfishness / motivation and irrationality.

Constructive Olympics?

I tend to have a recurring thought about this topic whenever the Olympics cycles around: whilst I am awed by the dedication that Olympic athletes exhibit in pushing themselves to the pinnacle human fitness and skill (in narrow disciplines), I don't consider the particular areas of focus to be all that interesting. Although there are no doubt useful derivatives of sports technology, in general Olympic athletes are not advancing humanity.

I wonder whether there are any people who engage in more intellectual disciplines that train as hard as and push themselves as hard as Olympic athletes? What happens if you take the most inherently mathematically gifted people in each country and then train them in the same way Olympic athletes are trained with: coaches; equipment; optimized diets; and (importantly for non-athletic endeavors), appropriate physical training?

Saturday, 6 August 2016

Home 3D printing

I suspect that rises in distribution network efficiency (particularly in urban/suburban locations) due to automation/drones are likely to outpace improvements in the flexibility and quality of 3D printing, suggesting that the future for at-home 3D printers is limited.

If Amazon stocks just about every conceivable item, and it can be delivered to your home by drone within an hour, do you really need a 3D printer?

Even if you a maker that uses a 3D printer to prototype items and make components, it may still be more cost-effective to use an online 3D printing service that delivers that 3D printed objects to your home, again via drone and probably within an hour of completion of the print job. Such 3D print services would likely be able to afford better 3D printers, and to upgrade them / replace them more frequently.

I believe that at home 3D printing will play an important role in those locations without access to a rapid distribution network.


Self-sustaining underground city

One of the reason's cited by those advocating the colonisation of Mars or the Moon (or orbital, or Venus) is to provide a refuge for humanity in the event of any catastrophe on Earth. A couple of quotes from Elon Musk on the subject are reproduced below (from the Guardian):

"It's sort of a futuristic version of Gibbon's Decline and Fall of the Roman Empire. Let's say you were at the peak of the Roman empire, what would you do, what action could you take, to minimise decline?"

"The lessons of history would suggest that civilisations move in cycles. You can track that back quite far – the Babylonians, the Sumerians, followed by the Egyptians, the Romans, China. We're obviously in a very upward cycle right now and hopefully that remains the case. But it may not. There could be some series of events that cause that technology level to decline. Given that this is the first time in 4.5bn years where it's been possible for humanity to extend life beyond Earth, it seems like we'd be wise to act while the window was open and not count on the fact it will be open a long time."

Whilst I entirely agree with doing what we can preserve intelligent life and civilisation, and I am keen that we colonise the solar system, I wonder whether there is a more cost-effective way to create such a refuge here on Earth.

It should be possible with current technology to build a self-sustaining underground city on Earth (or potentially many). Powered by geothermal energy, the city could be a closed ecosystem that recycled all air and nutrients. Food would be grown hydroponically.

In addition to recycling resources it may be possible for such a city to undertake a self-sustaining mining operation (by self-sustaining I mean that enough resources of the right types are extracted to maintain and create additional mining equipment). This would allow the city to slowly increase the living area to support population growth, and to develop further geothermal power sources to increase resource availability.

The city would be connected to the rest of humanity via the Internet, but would otherwise not be connected except in the event of a disaster in the colony. It is important to maintain this separation such that a pandemic affecting the surface world would not affect the underground world.

Advantages over a Moon/Mars/orbital refuge:

  • Proximity to Earth's surface for construction (particularly given the enormous cost of transporting material off the Earth's surface)
  • Proximity to Earth's surface for recolonisation
  • Proximity to Earth's surface for assistance in the event of disaster in the colony
  • Earth gravity
  • No additional latency in communicate with Earth's surface

Disadvantages compared with a Moon/Mars/orbital refuge:

  • Could potentially suffer from some of the same catastrophes that affect the Earth's surface (e.g. if directly targetted by a nuclear weapon)
  • Potentially lower availability of energy / resources
  • Separation from the economy of Earth's surface would be artificial
  • The proximity to Earth's surface would make such an endeavor feel less like a colony, with members (particularly second and later generation members) wanting to abandon the colony and rejoin the population on the Earth's surface

Thursday, 28 July 2016

Sharing space, power and network connectivity for science

There are already several projects that let you donate your spare computing cycles (and power and network connectivity) to science, notably the SETI@home and Folding@home projects. But what other ways can citizens support science passively from their homes?

An interesting expansion of this would be for citizens to host networked sensors at their homes that collect data from scientific research. There are several possibilities:

  • weather - air temperature, air pressure, rainfall, humidity, light sensors, wind speed sensors, etc
  • air quality - already done, see Air Quality Egg, TZOA, AirBeamLapka PEM, Clarity, CleanSpace
  • telescope - it is well within the capabilities of current technology to have an automated telescope that scans the skies capturing data with digital sensors and relaying that information to scientists via the internet. Potentially the telescope could be directed remotely, for example to track a comet.
  • biodiversity monitoring - cameras (both visible light and infrared), combined with image recognition technology could potentially be deployed to identify and count wildlife, particularly birds, and ideally indicator species
  • traffic sensor - a camera combined with image recognition technology that monitors the traffic in the road outside the house
  • soil moisture / water table level - this sensor would need to be dug into the ground, and would provide data on drought conditions and the risk of flooding
The key attributes of these systems is that they would be fully automated. All the citizen would need to do would be to purchase the device, site it somewhere appropriate (e.g. telescope on the roof) connect it to a power source and connect it to their internet connection.

Perhaps slightly more invasive to privacy, but nonetheless useful for psychological and sociological researchers, would be to install sensors to monitor human behavior in the house. However, such data collection is more closely linked to the individual that the location, and is already to a large extent covered by the quantified self movement (albeit not necessarily with the data shared with scientists).