Monday 22 April 2013

Automating datacentres

Do datacentres lend themselves to automation? In some respects yes. Racks and servers are standardised sizes. Rows are laid out systematically at standardised distances. However, wiring can be fairly tricky (even for a human, with superior image recognition abilities).

So how might datacentre automation work? Specially designed servers that can be racked and de-racked by a robot are designed and deployed. These would be similar to blade servers to avoid wiring issues. Robots would take servers from a store and place these into the racks. When a server fails, or the server has a component that fails, the monitoring system would give the robot the row, rack and server positions. The robot would then remove this server and deliver to a workshop for fixing or recycling. Meanwhile the robot would deploy a replacement from stock.

A similar approach (although slightly more complicated) should be possible for drives swaps of storage arrays. Again, some customisation of the design of the IT hardware would make the task easier for the robot.

This should be possible with existing technology levels, e.g. using floor guides (like Amazon's Kiva), and other optical cues.

It might also be possible to combine this idea with my thoughts on datacentres and lasers.

DNA degrading coating

Problem: surfaces are contaminated with the DNA of too many people, making identification of individuals associated with a crime difficult.

Potential solution: coat surfaces with DNA degrading enzymes that will reduce the half-life of DNA on the surfaces. As a result, only the most recent DNA will still be present, making identification easier.

Pressure sensitive kitchen worktop

Problem: when cooking, having to get the scales out to weigh out quantities. Incurring the extra washing up of the scales' pan.

Solution: pressure sensitivie kitchen worktop that allows any part of the work surface to be used as a set of scales, without clearing the worktop. To use the scales, the user would place a container on the work surface and hold the container of food to be weighed out (e.g. a bag of flour). The user would press a button on a small display on the wall behind the surface (or another convenient location) to turn on and zero the work surface. The user would then poor the food into the target container, watching the display indicate the mass transferred.

Possible problems: spills will count towards the weight measured.

Tuesday 16 April 2013

Entrepreneur Top Trumps

As we live in a world of entrepreneurs (yes, I'm jealous) wouldn't it be fun to have a game of Top Trumps based around entrepreneurs?

The variables could be the factors that made the entrepreneur successful, e.g.:
  • Hard work
  • Intelligence
  • Education
  • People skills
  • Right place/right time
  • Risk appetite
  • Resilience
Alternatively they could indicate the relative success:
  • Net worth
  • Age at first million $
  • Age at first billion $
  • Countries operating in
  • Market share%

Tuesday 9 April 2013

Compare stock price to subsequent series of dividends

I know it's an obvious analysis that's been done to death, but it would be interesting to me to see how the stock price divided by its subsequent series (length dependent on payback period) of dividends (discounted at appropriate discount rate) varies over time.

Based on the input criteria, i.e. payback period and discount rate, a value of greater than 1 means the stock was over valued, and a value of less than one means a stock was undervalued.

What proportion of the time is spend under or over valued? How has this changed over time? How do different stocks, different sectors compare? etc

Cloud audit-trail / control evidence service

The Problem: organisations need to maintain an effective system of internal control to ensure effective operation. To ensure and prove that controls are operating, evidence must be retained. Much of this is paper based or electronic, but due to the abilities of printers and image manipulation software, proving the date that the evidence was prepared, and proving who actually prepared it, become a problem.

The Solution: a cloud-based control evidence hosting service that integrates with key business systems. This solution includes user access control (proving the "who") and time/date stamping (proving the "when"). As it is a third party, the risk of manipulation is significantly reduced. Obviously, all information communicated over the internet would need to be encrypted at source, and stored in an encrypted form.

In an ideal world there would be multiple cloud-based control evidenced hosting services, such that each of these services could host their own evidence with an independent third party.

Psychology of group photo composition

Hypothesis: people appearing in the centre of the photo are perceived to be more important.

Experiment: assemble a group of people, and photograph them in every possible combination. Ask respondents to rank the importance of the people in the photo, with different groups of respondents getting different photos to look at. Control for the specific people (as tall people are perceived as more important). Look for patterns.

It would also be interesting to explore whether there are any cultural differences in the findings.

Monday 8 April 2013

Working 9 to 5....

...that's no way to make a living.

Observation: the relationship between the number of hours you work and how "hard" you work is non-linear. This observation ignores presenteeism (which itself heavily distorts the how "hard" people work). To explain, imagine that your job is to assemble cardboard boxes, and each box takes 30 seconds to make, that if you work at 120 boxes per hour consistently, the impact of that upon you physically, emotionally and intellectually increases per hour. Human beings have a finite capacity for productive output within a 24 hour period, requiring rest and sleep (and a break from monotony).

Hypothesis: the graph of "hard" vs hours per day looks something like the following:

The peak at 1h is recognition of the fact that, due to the human condition, it may be harder for people to motivate themselves to do one hour's work per day than to do two.