Most of us recognise and respond to threats subconsciously and do so very fast.  However not all threats can be perceived that way.  Some threats are more long term.  Their perception requires careful weighing up of evidence using intellectual instead of instinctual processes.

I’ll call  these threats and threat detectors type A and type B.

While type A threats need to be dealt with immediately, with type B threats there is time to for those who perceive them to communicate both the threat and possible ways to avoid it to others.

As individuals type A threat detectors will have a better chance of survival in a hostile environment with many immediate threats such as predators.  However groups that have some type B threat detectors will survive better than those that don’t in a hostile environment that also has longer term or periodical threats such as floods, fires, droughts etc.

Both types do better in a group that contains both than on their own.  In a group they protect each other from those threats that they are individually not so good at protecting against.

Threats exist not only from the external environment but also from others in a group.  While type A will tend to recognise and act on those threats at an unconscious or instinctual level, type B will tend to recognise and act on those threats consciously.  Such behaviour by type B being intelectual has to be learnt.

At least to the extent that social behaviour is the avoidance of threat in a social setting, type A will instinctively know how to act, whereas type B will have to learn how to act in each new social situation.

Given that type B threat detection is an intellectual activity, those with higher intelligence will tend to be better at it.  As a consequence type B threat detectors will have evolved and tend to be more intelligent than type A threat detectors.

With the range and types of threat that we have had to deal in our evolutionary past those groups with mainly type A and a few type B threat detectors have survived over those with more type B threat detectors.

Here I’ll make a leap that I am not qualified to make, but I’ll do it anyway.  On the basis that  type B threat detectors have to learn social behaviour and tend to have above average intelligence, I posit that type Bs are currently diagnosed or labelled as having Asperger’s syndrome.

Given the current nature of the threats that we currently face the world would be better of if more of us had Asperger’s.

I’d welcome any feedback on the above.

One of my distractions for the past couple of years has been playing Series of Tubes. This is a networking puzzle gadget by Wei-Hwa Huang that can be be added to a Google personalised home page.

Series of Tubes can also be played at
http://weihwa-puzzles.appspot.com/gadgetpage?g=series_of_tubes.

The Pieces – head, cross, tee, elbow & straight

The Series of Tubes game board consists of a number of squares in which there are tubes [, , & , ] and heads []. There are four types of tube – a cross [] which connects to the four adjacent squares, a tee [] which connects to three adacent squares, a straight [] and an elbow [] which both connect to two adacent squares.

Make all the users happy!

At the start the pieces are in random positions and one (tube or head) is coloured green.

Any piece that is rotated so that it is connected to a green piece also turns green.

The object of the game is to “Make all the users happy!” That is to rotate pieces until there is a pathway between all tubes and they all turn green.

Options – wrapping, size, marking yellow

Series of Tubes also has a number of additional options or features.

By default the pieces at the edge of the board only connect to adjacent pieces. For example any straight piece at the edge of the board has to run along that edge (as in the completed puzzle above). However there are options that allow for games in which the top edge of pieces at the top of the board can join to the bottom edges of pieces at the bottom of the board, and similarly for left and right edges. This is called wrapping.

The size of the board can also be altered. My current favourite is a 25 by 25 board with wrapping on both top-bottom and left-right.

As an aid to solving the puzzle, the background of individual pieces can be turned yellow.

Strategies

While my current favourite game is 25 by 25 with wrapping, I’ll illustrate strategies on a 10 by 10 game with wrapping.

Identify “fixed” pieces

There are four situations in which the correct orientation of pieces can be determined without reference to the orientation of any other piece.

Crosses are fixed

Crosses can not be rotatated, so their orientation is fixed.

Straights perpendicular to two heads

All pieces must be joined to all other pieces, so an arrangement of pieces consisting of one or more straight pices with a head at each end is invalid. Consequently if there are two heads with only (one or more) straight pices between them, then all of those straight pieces must be perpendicular to to the line joining the two heads.

Head with heads on 3 sides

The rotation of any head that is surrounded on three sides by other heads is fixed.

Combination heads & straights

The fourth case is a combination of the second and third cases. The rotation of a head is also fixed if it is surround on three sides by heads or by straights that are fixed because they are between the head in question and another head.

Rotate adjacent pieces

Once pieces who’s orientation is fixed purely due to their nature or location on the board has been determined, it is possible to determine the orientation of some adjacent pieces.

Straights adjacent to fixed pieces

The orientation of any straight adjacent to a fixed piece can be determined. It is either:

1. parallel to the fixed piece if that piece has no opening adjacent to the straight, or
2. perpendicular to the fixed piece if the fixed piece has an opening adjacent to the straight.

Adjacent fixed pieces with matching number of openings

If the fixed pieces adjacent to a piece have the same total number of openings on their adjacent sides as that piece has, then the orientation of that piece can be fixed. That is the orientation of:

1. a head adjacent to a fixed opening;
2. an elbow (or straight, if not already fixed by application of the previous strategy) adjacent to two fixed openings; or
3. a tee adjacent to three fixed openings

can be fixed.

Semi fixed elbows

The orientation of an elbow adjacent to one fixed opening is semi-fixed in that it can not have an opening opposite the fixed opening. Consequently any strategies that require no opening can be applied to the piece on the opposite side to the fixed opening. Equally any tee adjacent to wall (a fixed piece with no opening) is semi-fixed in that it must have an opening opposite the wall. Consequently any strategires that require an opening can be applied to pieces on the opposite side to the wall. This strategy can also be applied to a chain of adjacent elbows.

Tee with back to wall

Any tee adjacent to a wall, that is a fixed piece with no opening facing the tee must face away from that fixed piece.

Fixed elbows

Any elbow that is in a corner between two fixed pieces can be fixed. It will either be between:

1. two openings, in which case it will face the two openings; or
2. two walls, in which case it will face away from the two walls; or
3. a wall and an opening, in which case it will face the opening and away from the wall.

Adjacent fixed openings

Where there are adjacent openings on fixed pieces and those openings are joined by a fixed path, then the orientation of the two pieces adjacent to those openings can be fixed. Elbows and tees will face way from the other opening, because if the faced the other opening they would join (if the other piece were an elbow or tee), or leave an un-joinable opening (if the other piece were a stright or head). This is because loops and open ends are not permitted. We alrady have strategies for heads or straights adjacent to the openings.

Corridors

Once a fair number of pieces have been fixed, corridors of unfixed pieces one piece wide are sometimes formed. When they do, and there is an end to the corridor, then all the pieces in the corridor can be fixed. Also if there are two heads in the middle of an open corridor then they can be fixed (as they can not face each other) and every piece in both directions down the corridor from these two can be fixed. [This later case not illustrated.]

Repetition

In this example the solution can be found by repeating these strategies until all pieces have been fixed.

Other strategies

Other strategies [which were not needed in this example], include:

Multiple heads aginst a wall

Where several unfixed heads are in a line adjacent to a wall (fixed pieces with no openings), then the heads in the middle must face away from the wall.

Heads in a corner

Where two adjacent unfixed heads are against a wall, and one of the heads is in a corner between two walls, then the one in the corner must face away from the walls and the other head.

Semi-fixed elbow and two heads

When an elbow is semi-fixed to a head and one option is to join the other end of the elbow to another head, then this position is not valid, so the elbow must point in the other direction.

Adjacent fixed pieces with matching number of closed sides

If the number of closed sides of fixed pieces adjoining a piece is equal to the number of closed sides that piece has, then the orientation of that piece can be fixed. [This strategy was suggested to me by Lars Huttar. It is the opposite of the "Adjacent fixed pieces with matching number of openings" strategy (which is a generalisation of "Fixed elbows a.") and this strategy is a generalisation of "Fixed elbows b."]

Multiple Solutions

Some games have multiple solutions. In most cases this occurs when four or six adjacent pieces can be placed in different positions and still produce a valid solution (all users happy).

In other games a loop corridor will appear around a central island and the pieces in the corridor can face in either direction around the corridor to produce a valid solution.

Occasionally a larger number of pieces will not be fixed and there will be multiple possible solutions.

After a short while I started looking for anyway to setup Office 2007 with the classic User Interface that Office had had up until Office 2003.

My prayers were answered when I found pschmid.net’s Ribbon Customizer which comes in two versions, Professional and Free. Both versions include Classic UI tabs which you can add to Excel, PowerPoint and Word 2007.

Since installing the Free version I’ve been able to use Office again.

There is a useful post on Cnet News on Changing the default folder view in Vista. This in turn points to detailed instructions on HOW-TO: Disable Automatic Folder Type Discovery.

On investigation it appears that the problem is due to the use of post codes to locate places that are being searched for.

The Royal Mail Address Finder returns 15 addresses for HA3 6SH, the post code for Grim’s Dyke. One address is for Grim’s Dyke, the other 14 are for flats located in Portman Hall.

Guess where Google and most of the other map and hotel search services place Grim’s Dyke. Yes, at Portman Hall, which is located over a third of a mile to the south east of Grim’s Dyke. Those that put Grim’s Dyke in the wrong place include:

• maps.google.co.uk
• www.multimap.com
• streetmap.co.uk [can only search for the postcode]
• london-discount-hotel.com
• lastminute.com

There are two hotel sites that place Grim’s Dyke in the correct place, pointing to the hotel itself or the entrance to the hotel grounds:

• booking.com [strangely the map is powered by google], and
• www.hrs.com

There are three organisations involved in this stuff up, all of whom could and probably would way its not our problem. The organisations and their probable responses are:

• The search providers: “We rely on data provided by TeleAtlas (or similar company). If there is a problem take it up with them.”
• TeleAtlas (or similar): “We provide the best location for each post code, in this case the location of 14/15 addresses for HA3 6SH.”
• Royal Mail: “Any use of post codes for purposes other than the delivery of mail are not our problem.”

So how do booking.com and hrs.com get it right?

Future comments will have more substance and many will have links to sites and pages of interest.

Here is a link to the CSEM Home Page.

Untill the next time,

Michael.