An article from my old site that’s popular with the search engines. On dropping an iPod Shuffle into the toilet, and learning something as a result…

Long drop

After riding to work I used to get changed in a cubicle in the gents’. Unfortunately, one day I forgot that my iPod Shuffle was clipped onto the waistband of my Ronhill Bikesters, and on removing my T-shirt, the Shuffle pinged off, danced around the seat in one of those slow-motion moments immediately preceding disaster, and fell into the water.

Fortunately, the facilities had been cleaned the previous evening and didn’t appear to have been used in the meantime, and the device must have been under for no more than three seconds. I whipped it out of the water, shook it vigorously, and wrapped it in paper to take up any water from the gaps around the switches and from the socket.

Flat

Once the Shuffle had dried out for a couple of hours, I risked turning it on only to find the battery indicator in the red, sometimes not illuminating at all. Not a big surprise: the water had probably discharged it, I thought. Things were fairly hopeful, as it would still play music through the headphones.

The problem

At home, I plugged it into the iTunes PC (running Windows 2000 at the time, now running Ubuntu Linux), but it wouldn’t charge — the expected orange light didn’t come on. I tried all sorts of things, with an increasing concern that I’d trashed my first iPod after less than a month, eventually solving it by plugging the dock into the (Windows XP) laptop.

The solution

Armed with this clue, I found out this piece of quirky behaviour: to get an iPod to charge under Windows 2000, it must be enabled for disk use. The only disadvantages of this setup are:

1. you must manually disconnect the iPod within iTunes before unplugging;
2. you cannot see the charge status on the Shuffle, as the LED flashes orange to warn you not to unplug it without disconnecting.

Luckily for me (the Shuffle was a Christmas present from my wife) it works perfectly now. I guess my warranty with Apple has gone (…wait for it…) down the pan (!) as a result of writing this article.

It started with this photo. We were going for a nice walk around the village of Rowsley in Derbyshire, when I spotted this sign fixed to the front gate of a private house. Intrigued by its history, I took a quick photo and enjoyed the rest of the walk, followed by a nice cup of coffee and a cake at the Cauldwell’s Mill cafe.

After uploading the photo to Flickr, I noticed the very subtle LNER lettering at the right-hand edge of the sign, indicating its origin on the railways. A quick search established that there were a few of these signs about, so the obvious thing to do was to set up a group for them.

Thus was born the Forty Shillings Fine group on Flickr. Every so often, I do a quick search to find new pictures of such signage, and invite the photographer to add their image to the group. Maybe one day, someone even less popular at parties than myself will find a valuable use for this collection of photos…

I am not a trainspotter or railway enthusiast (other than having an appreciation of the engineering involved in early steam engines). I do not own a model railway and am not able to answer questions on engines or railways. If anyone has one of these signs they no longer want, let me know!

This photo is a derivative work of this shot by Ben Harper, used in accordance with the photo’s licence.

Referring to the tables in the music theory page if necessary, we determine that, for G#m7, these notes are G#, B, D# and F#. The (#5) in the chord name indicates that the fifth is to be sharpened a semitone to Dx (D double-sharp), which is the same as E on a guitar.

So how do we work out a finger pattern? Let’s draw a grid and put marks where notes in this chord lie:

   0   1   2   3   4   5   6
E  |---|-x-|---|-x-|---|---| E F# G#
B  |---|---|---|---|-x-|---| B E
G  |-x-|---|---|-x-|---|---| G# B
D  |---|-x-|---|-x-|---|-x-| E F# G#
A  |---|-x-|---|---|---|---| B
E  |---|---|---|-x-|---|---| G#

Here, an -x- marks a potential finger position. We want to ensure as many notes from the chord as possible are present: if notes are omitted, the chord can become ambiguous and not sound as intended.

Now we need to work out a compromise between easy fingering and including the important notes in the chord (arguably, all of them!). Looking at the grid above, the most obvious shape is this:

   3   4   5   6   7         
E  |(1)|---|---|---| G#
B  |---|-2-|---|---| E (=Dx)
G  |(1)|---|---|---| B
D  |(1)|---|---|---| F#
A X|---|---|---|---| 
E  |(1)|---|---|---| G#

Note that we start on fret 3 now, not the nut, and that the A string is not played. In the absence of a bass player, I’d probably pluck the bottom string with my thumb; if I had a bassist or someone else playing the low G#, I’d probably just strum the top four strings. As this does not use any open strings, it can be used anywhere on the neck for arbitrary m7(#5) chords.

That’s an extreme example of a very complex chord that you are unlikely to ever need—unless you need to play the Bee Gees’ I Could Not Love You More. Here are some simpler “real-world” examples:

Amaj9

The chord theory page tells us that this is root, third, fifth, major seventh and ninth, and conveniently provides the notes: A-C#-E-G#-B. The first thing to notice is that three of these notes correspond to open strings.

Let’s assume a five-string chord, starting on the open A string, as a variant of the standard A chord. As this could be simple, let’s work it out without the chart. Working up from the open A string, there’s an E on the second fret of the D string, a G# on the first fret of the G, and the B and E can be left open. That gives us:

   0   1   2   3   4    
E O|---|---|---|---| E
B O|---|---|---|---| B
G  |-1-|---|---|---| G#
D  |---|-2-|---|---| E
A O|---|---|---|---| A
E X|---|---|---|---|

Pretty easy to finger. The disadvantage is that it omits the third (C#), which gives the chord its major feel. In practice, this doesn’t really affect things. If you were really bothered about this, you could soon work out:

   4   5   6   7   8    
E  |---|---|-4-|---| B 
B  |-1-|---|---|---| E 
G  |---|-3-|---|---| C#
D  |---|-2-|---|---| G#
A O|---|---|---|---| A
E X|---|---|---|---|

This is a bit harder to finger, but includes all the notes.

Dmaj7/G

We’ll start from the basic open D chord that most people know:

   0   1   2   3   4    
E  |---|-2-|---|---| F#
B  |---|---|-3-|---| D
G  |---|-1-|---|---| A
D O|---|---|---|---| D
A X|---|---|---|---| 
E X|---|---|---|---|

Turning this into a Dmaj7 chord requires a C# (see chord theory page if necessary). It’s easy to see what to do here:

   0   1   2   3   4    
E  |---|-3-|---|---| F#
B  |---|-2-|---|---| C#
G  |---|-1-|---|---| A
D O|---|---|---|---| D
A X|---|---|---|---| 
E X|---|---|---|---|

OK—now we need to get a G in the bass. We could turn this into a three-string chord without the open D and with an open G string, but that would be a weak-sounding chord. The lowest G is the third fret of the low E string, although it’d be hard to get a finger there from the shape above. The solution is to use a partial bar:

   0   1   2   3   4    
E  |---|(1)|---|---| F#
B  |---|(1)|---|---| C#
G  |---|(1)|---|---| A
D O|---|---|---|---| D
A X|---|---|---|---| 
E  |---|---|-3-|---| G

The A string is not played here—I use my right thumb to pluck the string whilst strumming the top four with my fingers. The A could theoretically be played open, but the small interval (tone) between the low G and low A would give a very muddy sound.

Ebsus4

Eb is awkward with standard tuning. The standard E-shape for Eb major (without the sus4) would need the bar at fret 11, which is highly impractical for general use. The A-shape is at fret 6, so let’s try that.

   5   6   7   8   9    
E  |(1)|---|---|---| Bb
B  |---|---|-4-|---| G
G  |---|---|-3-|---| Eb
D  |---|---|-2-|---| Bb
A  |(1)|---|---|---| Eb 
E X|---|---|---|---|

The sus4 replaces the major third (G) with the fourth (Ab). There’s only one G in the chord above, so to go up a semitone is pretty easy.

   5   6   7   8   9    
E  |(1)|---|---|---| Bb
B  |---|---|---|-4-| Ab
G  |---|---|-3-|---| Eb
D  |---|---|-2-|---| Bb
A  |(1)|---|---|---| Eb 
E X|---|---|---|---|

An alternative for long-fingered players would be to turn the open D into a four-string Eb:

   0   1   2   3   4    
E  |---|---|-3-|---| G
B  |---|---|---|-4-| Eb
G  |---|---|-2-|---| Bb 
D  |-1-|---|---|---| Eb
A X|---|---|---|---| 
E X|---|---|---|---|

…and raise the G as above:

   0   1   2   3   4    
E  |---|---|---|-4-| Ab
B  |---|---|---|-3-| Eb
G  |---|---|-2-|---| Bb 
D  |-1-|---|---|---| Eb
A X|---|---|---|---| 
E X|---|---|---|---|

Cm6/Eb

This isn’t even close to a standard chord, so we’ll have to start from scratch. Cm6 is C-Eb-G-A, and we want the first inversion with the Eb as the lowest note.

Let’s mark out possible finger positions again:

   0   1   2   3   4   5   6   7   8
E  |---|---|-x-|---|-x-|---|---|-x-| G A C
B  |-x-|---|---|-x-|---|---|---|-x-| C Eb G
G  |---|-x-|---|---|-x-|---|---|-x-| G A C Eb
D  |-x-|---|---|---|-x-|---|-x-|---| Eb G A
A  |---|---|-x-|---|---|-x-|---|---| A C Eb
E  |---|---|-x-|---|-x-|---|---|-x-| G A C

The bottom E string is no use, as an Eb in the bass would be on fret 11. I’d try this for a bar chord:

   4   5   6   7   8    
E  |(1)|---|---|---| A
B  |---|---|---|-4-| G
G  |(1)|---|---|---| C
D  |(1)|---|---|---| G
A  |---|-2-|---|---| Eb
E X|---|---|---|---|

Again, as a bar chord, this can be used for any m6 first inversion.

A four-string version of the chord is:

   0   1   2   3   4    
E  |---|---|-4-|---| G
B  |-2-|---|---|---| C
G  |---|-3-|---|---| A
D  |-1-|---|---|---| Eb
A X|---|---|---|---|
E X|---|---|---|---|

Like the standard F chord, this is not a bar chord (no bar) but it also isn’t an open chord (no open strings), so it can also be used in any position.

This is a big table, but it’s quite logical and there are a lot of patterns (G7 is G plus dominant seventh; G9 is G7 plus ninth (octave-and-tone); G11 is G9 plus eleventh (octave-and-fourth); G13 is G11 plus thirteenth (octave-and-sixth)).

Self-test question 1

What notes are in the chord of C major?

Reveal…

Interval:          root   third   fifth
Semitone interval:  0       4       7
Note:               C       E       G

Self-test question 2

What notes are in the chord of F#m7?

Reveal…

Interval:          root   minor third   fifth  dominant seventh
Semitone interval:  0          3          7          10
Note:               F#         A          C#         E

Self-test question 3

What is E-G#-B-D#-F#?

Reveal…

British courts are being told to dole out harsher penalties for poor driving (PDF). In an interview primarily focusing on the cost of international text messaging, BBC Radio 4’s PM programme interviewed eminent linguist (and txt advoc8) David Crystal and discussed texting whilst driving which is explicitly called out in the new guidelines as being dangerous.

He said something along the lines of “it’s taken a few years for us to realize it’s dangerous to text whilst driving”. I’m certain this isn’t the exact meaning he intended.

The actual fact is that it’s taken us a few years to realize that there are people out there stupid enough to not realize that texting whilst driving is dangerous, and it therefore needs banning in law.

Juggling whilst driving is also dangerous, but there’s no explicit law against that — yet…

For convenience, we would set out from my grandparents’ house near Chichester, head to Portsmouth and catch the ferry to Caen / Ouistreham. From there, we would pick out the route that was the best compromise between direct and scenic. The aim was to get there as quickly as possible. First estimates on the map suggested about 720 miles from Ouistreham to Port Grimaud.

Equipment

We wanted to be as self-reliant (cheap!) as possible, staying at municipal campsites and cooking our own food. We’d have to carry all our own food, clothes and camping gear. I also wanted to take my “new” Olympus OM10 to take photos of the trip. Plenty of luggage space required!

I had an old Peugeot Super Competition racing bike adapted for light touring—36-spoke wheels, a Blackburn rear rack and a lowrider-type front rack. I removed the mudguards for the trip, but retained the Profile aero bars. Neil had a new Saracen mountain bike with rear rack; he fitted bar ends to give more hand positions.

Training

I had done quite a bit of cycling prior to this trip, but no long journeys for quite a while beforehand. In particular, this would be the first time I’d break 70 miles since a knee operation a few years ago. I did a four-day training ride about two weeks before the trip, from my grandparents’ in Chichester, stopping off with my then-fiancée Sam during her medical training in Salisbury, and continuing to my other grandparents’ house in Bristol before retracing my steps. Four consecutive 60-mile days, and I didn’t feel too bad. Encouraging.

Neil had also done long rides in the past, and prepared with a couple of longish trips around his home near Birmingham. Once he came down to Chichester, we tried a 30-mile ride with the bikes loaded up to make sure everything stayed on.

The ride

All my photos from the ride are on Kodachrome 64, taken on an Olympus OM10 with a Sigma 28mm f2.8 lens and a polarizing filter (often overused—I was a beginner!); and scanned in, some using a Minolta Scan Dual IV, some with a cheapo UMAX flatbed scanner, compressed to suit low-bandwidth modem conncetions that were common when I first published this article on my old web site. Quality is generally terrible, and now I’m more into photography, I’m ashamed of these shots!

Day 1: Chichester to Caen

HMS Britannia at Portsmouth

Neil at Portsmouth

Me at Portsmouth

This was a bit of a warm-up day, riding from home base to the port as Portsmouth, then spending nine hours on the ferry, then riding to a campsite just outside Caen before dark. Saw HMS Britannia from the ferry. Total ridden distance was just under 50 miles.

Whilst waiting for the ferry, a couple on a tandem started talking to us, then started criticizing our bikes, saying we’d never make it on one bike with racing tyres (23mm) and one with tractor tyres (1.5″ slicks)! It turned out they were just pottering around Brittany for a couple of days. If that was you: we made it easily with almost no trouble—see below.

Day 2: Caen to Sées

A very boring day—fighting our way south on big rolling dual carriageways. At least we had a tailwind and sunshine. Neil was suffering from a stomach upset, which made his day even worse.

Day 3: Sées to Vendôme

Bad start to the day—about 10 miles out of Sées, I thought I’d left my passport behind at the site. I was just about to turn back for it leaving Neil to wait when an intensive search of my panniers uncovered it.

Better roads and scenery. Discovered that France closes for lunch, so if you want to eat, buy in the morning! In Vendôme, we’d stopped to look for the campsite, and a friendly Frenchwoman in a Fiesta drove up and asked if we were OK. I explained we were looking for the campsite, and she beckoned us to follow her. 30mph through the suburbs after 85 miles isn’t funny; she did get us to the site though. Friendly Belgians gave us some chocolate. After a day’s hot cycling, half a bottle of cheap table wine each was enough to make us very unsteady on our feet! Total distance: 87 miles.

Day 4: Vendôme to Vierzon

Blois from the bridge

Sunny with some clouds, tailwind still going strong. Very pleasant day, including crossing the Loire at Blois. Nice campsite by the river; hot showers! Phoned Sam to let her know we were still alive and enjoying ourselves.

Day 5: Vierzon to Montluçon

Basilica at Chateauneuf-sur-Cher: interior

Basilica at Chateauneuf-sur-Cher: exterior

Today we had the only headwind of the ride, crossing open cornfields. Plenty of slipstreaming. Hot sun and plagues of locusts (or maybe just crickets) all over the road.

I had flat tyre problems on this day—not punctures, just an old inner tube letting go. Tried to repair it a couple of times, but had to replace it in the end. Other than a dried-up chain needing oil, this was the only bike problems we had.

About halfway through the day, France started to get a bit hilly. Nice but nondescript ride. Campsite at Montluçon was horrible, thanks to large numbers of “gypsies” (not true Roma) making lots of noise all night and being generally messy.

Day 6: Montluçon to Thiers

Neil emerging from the tent at Montluçon

Started with a steep hill out of Montluçon. Scenic but uneventful. Nice campsite near Thiers—took a little while to find, though! Discovered that even McDonalds don’t sell a substantial breakfast in France.

Day 7: Thiers to Solignac (Le Puy en Velay)

Resting on the road to Ambert

Eglise Saint-Michel d’Aiguilhe, Le Puy en Velay - from this site

Neil’s hands were giving him trouble—the lack of riding positions on the mountain bike was playing havoc to the extent that he couldn’t touch his fingers together. Lent him my gel-padded mitts which seemed to help a bit. The problem lasted for several months after the ride.

The day ended with a descent into Le Puy en Velay past the very bizarre church on the rock, then the shock of finding that the campsite at Solignac was up another huge hill. Yellow Michelin maps don’t show terrain too well…

Day 8: Solignac to Lanas (Aubenas)

Solignac-sur-Loire; Chadron; Saint Martin de Fugeres; Salettes; Aubenas; Lanas

Viaduc de la Recoumène near Chadron

Alpine view

Neil enjoying the Alptastic experience (top of hillock)

Pleasant rural start. Very strange huge curved brick bridge out in the wilds near Chadron (below, map); couldn’t work out what it was for. It turns out to be the Viaduc de la Recoumène, built in 1925 as part of an unfinished railway project called La Transcévenole: more information at Structurae. Thanks to Valérie Valette for this information!

Lots of climbing—first time over a mile high on a bike. Met a German cyclist who’d been biking around Europe for 3000 miles and planned to finish that day, meeting his girlfriend in Aubenas.

In the afternoon, we came down off the Low Alps onto the coastal plain. This descent was the most fun I have ever had on a bike—perfect road surface, nearly a mile of vertical descent over about twelve miles of road, cruised at over 30mph most of the time. We named the experience “Alptastic”! Stopped a couple of times to admire the view and let our rims cool down.

Halfway down the descent, we met the German, stopped with a despondent look on his face and a flat rear tyre. “It is finished”, he said. We thought he was being a bit melodramatic—was this his first ever puncture?! Then he pointed to the section of his rear wheel where the brakes had worn through the rim… He had phoned his girlfriend and she was coming to pick him up, some ten miles short of his destination after several thousand miles.

View towards Aubenas

Aubenas nearly killed me. It was very hot with stifling exhaust fumes: after the clear air of the mountains, I felt very ill indeed. Then I ate 3½ choc-ices for lunch (Neil only wanted 1½, and we’d bought a packet of five!) which made matters much worse. I don’t remember much of the ride to the site at Lanas.

Day 9: Lanas to Aix-en-Provence

Lanas; Pradons; Ruoms; Vallon Pont d’Arc; Salavas; Barjac; Saint-Gervais; Bagnols sur Céze; Avignon; Cavaillon; Aix-en-Provence

An awkward decision—we had 190 miles to go, and we felt like we wanted to do it in two days. However, we didn’t want to arrive at the campsite too late, and Neil knew that there was a big ridge before the coast. We decided to try to put in a long day and push for Aix-en-Provence, some 115 miles away, leaving us a mere 75 miles for the final day.

Vallon Pont d’Arc

Getting hot before Avignon

Avignon

This was probably the most scenic day to start with: the Ardèche valley was really pretty, and we crossed the river at Vallon Pont d’Arc. Stopped in Barjac for the mid-morning break to find a large bustling market. Popped into a supermarket for more choc ices.

We had the only rain of the trip on this day: coming into Avignon on the main road, it started to get really hot and oppressive. We stopped in Avignon for lunch, and the heavens opened. Had to shelter for about twenty minutes, then it all stopped. The next section was all industrial main roads, and was very unpleasant.

We made it to Aix, covering 115 miles, and descended into the town at over 35 mph. The woman at the campsite asked where we had come from that day, and would not believe our answer of “Lanas, near Aubenas”. She made me write it down—turned out she commuted from Aubenas every day and couldn’t believe anyone would cycle it.

Day 10: Aix-en-Provence to Port Grimaud

Aix-en-Provence; Saint-Maximin; Brignoles; Besse sur Issole; Vidauban; Les Arcs; Le Muy; La Garde Freinet; Grimaud; Port Grimaud

About to cross the final ridge

Made it!

The beach at Port Grimaud

Port Grimaud village by night

Liberation Day fireworks

This day was a breeze—a relatively low mileage with the knowledge that this was the last day. Simple route, too: we just followed the route nationale that runs parallel to the autoroute as far as Le Muy, then turned south over the big hill through La Garde Freinet (which sounds as though it’s a warning about brakes, aptly).

Neil found a burst of energy and I struggled to keep up over the ridge. Very hot day—we stopped for a drink at La Garde Freinet before the exciting descent through the old village of Grimaud.

Arrived at the site at about 3pm, having covered 74 miles. Family not yet arrived: we met up with some family friends and ate chips. Sam and parents turned up the following day, surprised to find us there already. We had covered a total of 720 miles in France (i.e. not counting the Chichester to Portsmouth to Caen mileage) over nine days—an average of 80 miles per day.

ZX Spectrum (Bill Bertram)

When I was just the right age for that sort of thing, the home microcomputer emerged on the market. Sinclair was the chief driving force in bringing affordable computers into the house, starting with the ZX80 and ZX81. My first home computer was a Version 1 ZX Spectrum: we bought the 48K (yes, kilobyte!) version rather than the 16K, as my father had previously run into memory limitations writing a protein simulation program on a 32K BBC B for the school he taught at, and understood the value of RAM.

I also used a BBC B, and had experience with friends’ computers, machines like Acorn Atom, Oric-1 and the Commodore 64 and VIC-20. All of these were radically different machines compared with today’s homogeneous offerings.

As home computers were so basic and so limited at the time, it was important to fully understand the internal workings (both hardware and software) if one wanted to do anything useful with them. Here’s a random list of things I had to get my head around as a kid, when learning and understanding is at its best:

• the difference between RAM and ROM;
• BASIC, assembly language and machine code;
• compiled versus interpreted languages;
• Z80 machine code;
• memory-mapped graphics (where the screen is simply a representation of a section of user RAM, and writing to this memory directly affects the display);
• tape storage and audio data interface;
• serial and parallel interfaces;
• control flow optimization;
• binary and decimal.

I thought it worthwhile jotting down a few memories so that those of a similar age can reminisce, and those pesky kids on my lawn who listen to that terrible noise they call “music” can look in befuddled incomprehension.

• Usborne books with robots running around the inside of the computer, delivering 0s and 1s into a row of little boxes;
• “thermal” printers costing six times more than modern colour inkjets in real terms, printing smudged black characters onto silver paper rolls;
• building user-defined graphics characters as a string of eight bytes;
• fighting for use of the television set: no dedicated monitors for us back then!
• counting each clock cycle in a machine code program;
• typing in a long BASIC program out of a magazine, only for someone to accidentally knock the power lead out of the machine, losing hours of laborious work;
• 45rpm floppy records on magazine covers: plugging the Spectrum into the record player to load up programs;
• program distribution over the radio;
• writing a full-screen video player. The executable was just 11 bytes long and would copy pre-stored frames from the user RAM to the screen memory one after the other. I forget the frame rate, but it was fast. Sadly the RAM could only store six frames, and the graphics system could only do two colours per character space.
• wearing out key matrices playing Daley Thompson’s Decathlon too much: replacing them a couple of times and eventually building a custom box, PCB and keyboard;
• working through all 256 screens on Sabre Wulf while my sister drew a map of the maze;
• learning about pirating (to which I’m now vehemently opposed) and circumventing copy protection. A friend and I once copied the reference card for Jet Set Willy by hand with four felt-tip pens. In today’s world of digital cameras and the Internet, such a copy protection scheme would last about 1/250s in good light.
• learning CAD/CAE fundamentals from VU-3D and Make-A-Chip.

Wow. Now I feel old.

Spectrum owners should visit World of Spectrum for a trip back to the past. Spectrum emulators let you play some of those old classic games legally, thanks to Amstrad and WoS on your modern computer. I must warn you, though: most aren’t as good as you remember…

Anyone who missed this nascent period in home computing missed out on a valuable learning experience, I believe. Some of those who were there went on to start the demo scene, which was undoubtably responsible for many advances in computer graphics. Next time you watch a Pixar film, for example, you’re probably benefitting from the work of people who started out with ZX81 code like:

This house had a real air of magic about it to me as a small child — it was quite a large house, detached with five bedrooms, set in just under an acre of land. My grandparents had lived there since the house was built in the 1930s, and much of the furnishings and decoration was old. The house was filled with little mementos from their trips abroad: not today’s plastic tat, but glass, shell, porcelain, wood, metal and so on. There were little lamps made from miniature liqueur bottles and glass paperweights. The house had a particular smell which I’ve not experienced anywhere else.

It had a coal shed, and bellpushes in two of the bedrooms and in the lounge to summon the maids that would have been commonplace when the house was built. The boiler that provided heat and hot water was coke-powered, and various generations of cat could be found sitting on or around it in cold weather. Shelves containing firewood, woodlice and spiders lined the alcoves at the sides of the porch. The whole house gave an air of self-sufficiency through old-fashioned necessity rather than modern eco-friendliness.

The garden was equally as magical to a young child as the house. An orchard at the rear of the house provided apples and pears from about twenty trees: the fruit was usually harvested off the ground, with the risk of turning an apple over to find it half-eaten and full of inebriated wasps. A vegetable garden featured runner beans, purple sprouting broccoli, potatoes, carrots, spring onions, sprouts, asparagus, raspberries, gooseberries and many other crops. A large concrete-framed greenhouse contained tomatoes, which gave off a heady aroma, and grapes grew from a vine on the roof strut. Trees and dense hedges gave plenty of hiding places and climbing opportunities, and fallen brown holly leaves taught the folly of skulking around barefoot.

There was also the “little” lawn, nomenclature that was definitely relative, which doubled up as a nine-hole miniature putting green; a grass tennis court and the front lawn with an old caravan and two hammock-spaced willows that were frequently put to that use.

The outbuildings were also great for me as a youngster: the garage housing the old green Morris Oxford was dark, with mysterious things stored in the roof space. A workshop with an old circular saw table and a workbench came off that — I spent much time in there building and fixing things, making use of the huge variety of ironmongery stored in metal tobacco tins.

The woodshed contained at least three mowers: two cheap and past-it electric mowers, and the petrol-powered “new” mower, which was only new relative to some ancestral piece of machinery that presumably expired long before I was born. This was self-powered and could pull a home-built trailer on which the operator could stand. I often used this when mowing the little lawn and tennis court, from well before the age of ten. Many other exciting things could be found in the woodshed: a large metal drum on a shelf containing petrol for the mower and imbuing its own distinct aroma on the interior; ancient oilcans with a thumb-pump; a 1966 E-reg Honda C90 scooter which I never saw in use; many old bicycles, a petrol rotavator and seasonally, strings of onions tied up with red baling twine drying out.

The house was set in the corner of a field. We would often have bonfires to get rid of garden refuse at the edge of the field, some of which would smoulder for days. The farmer would leave bales of straw around after the harvest, and my uncle would “help” me build bale houses (in reality, he must have done pretty much all of the work, but you know how kids are).

The house is just over a mile from the sea and on a flat peninsula, so we had many trips to the beach, mostly by bicycle. One of my projects was to build a trailer allowing me to pull our large sea-canoe down to the beach behind my bike.

The house is still in the family, but the magic has all gone and it seems a lot smaller now. I don’t think it’s down to the refitted kitchen or the redecoration; I think the magic was self-generated, by the excitement and awe of childhood.

Where am I going with all this?

That house and its feeling is what comes to mind whenever I read John 14:

1Do not let your hearts be troubled. Trust in God; trust also in me. 2In my Father’s house are many rooms; if it were not so, I would have told you. I am going there to prepare a place for you. 3And if I go and prepare a place for you, I will come back and take you to be with me that you also may be where I am. 4You know the way to the place where I am going.

When Jesus spoke these words, a “house with many rooms” would have been something special, out of the ordinary. Nowadays, comfortable, multi-room houses are the norm and these verses lose some of their impact as a result.

One day, when I am called home, I will be back in a place with the same feeling of awe as described above, but on a much larger scale — and that will never wear out.

I’m looking forward to it! Will I see you there?

That gets you a full-featured HDR construction and tone-mapping toolkit. When you first try to build an HDR image from multiple LDR images, you’ll see an option to align images using hugin’s align_image_stack:

This utility is not available in any of the stable releases of hugin at the time of writing. There are instructions on compiling align_image_stack from source, but this does not cover the large number of dependencies that cmake and hugin drive.

Never fear: I have already jumped through all the hoops for you, ensuring that within a couple of hours of writing this, a new version of hugin including align_image_stack will be made available. Here’s what you need to do, for Ubuntu 8.04 (Hardy Heron LTS):

This process takes a surprisingly long time: the utility depends on much of the hugin code. Once finished, fire up qtpfsgui once again, and you should be able to use align_image_stack to align those slightly misaligned source images. Anecdotal evidence suggests that you may get more reliable results running it from the command line, then opening the resultant HDR file in qtpfsgui.

If it still doesn’t work for you, you could try hdrprep instead. Please comment on your experiences with either method!

   0   1   2   3   4   5   6
E  |---|---|---|---|-4-|---|  A
B  |---|---|-2-|---|---|---|  D
G  |---|-1-|---|---|---|---|  A
D O|---|---|---|---|---|---|  D
A X|---|---|---|---|---|---|
E X|---|---|---|---|---|---|

The reason power chords are used more extensively in rock, usually with distortion, is this:

A note put through a distortion unit will sound like a major chord already—adding the fifth is acceptable, but adding the third destroys the chord.

Let’s look at why this is. We need to cover harmonics, scale theory and the properties of distortion units to do this. Here we go…

Harmonics

Harmonics are the multiples of any fundamental note. For example, an open A string at 110Hz will have a second harmonic at 220Hz, a third at 330Hz, a fourth at 440Hz, a fifth at 550Hz and so on.

On the guitar, you can produce the nth harmonic by lightly touching the string at 1/n of its length and plucking it. Touch a string very lightly above the twelfth fret and pluck it—you get the second harmonic, which is exactly one octave higher (see below). Doing the same above the seventh fret gets you the third harmonic, which is almost exactly an octave-and-fifth. You can also touch the string at any of the n equally spaced points—so you could touch at 2/3 the length, over the nineteenth fret, and get the same note. Of course, touching at 2/4 of the length will get you the second harmonic not the fourth (as 2/4 ≡ 1/2).

You can demonstrate what is happening with a rope. Get someone to hold one end. You hold the other and start to swing the rope like a skipping rope at the frequency that feels most natural. Now try doubling, tripling or quadrupling that frequency. See the stationary points (nodes)? Back on the guitar, your damping finger forces the string to stay still where you are touching it. The string will vibrate at the lowest harmonic that has a node at that point.

Scale theory

The common scale in Western music is made up of octaves (doublings of frequency) each divided into twelve equally-spaced semitones (although read up on tempering). The ratio between each semitone is therefore the twelfth root of two (hereafter referred to as r) which is about 1.059.

Distortion units

Solid-state distortion units tend to add odd nth harmonics (third, fifth etc), decreasing fairly rapidly as n increases. This distortion tends to sound harsh. Valve distortion tends to boost the even harmonics, which sound warmer.

Relationship between harmonics and notes

OK—so what note does the nth harmonic relate to? We need to find out how many semitones correspond to a multiple of n. This is:

x = log(n) / log(r)

…and so on. Usually, anything above the fifth harmonic is quiet enough as to solely form part of the sound of the instrument, distinguishing it from a sine wave, rather than being recognizable as a distinct pitch.

Using this, we see that a distorted power chord of C (contains C and G) will generate third harmonics of G (from the C) and D (from the G) and fifth harmonics of E (from C) and B (from G). This gives a power chord the feel of an unbalanced Cmaj9 chord (C-E-G-B-D). Adding your own third would muddy this further: here, a major third of E would add harmonics corresponding to another B and a G#. Alternatively, a minor third (Eb) would add a Bb and a G. Both of these would result in a semitone clash (G-G# or Bb-B).

I’m fairly new to Movable Type on which this blog is based, but I am familiar with several other templating languages. I wanted to provide a link to the previous entry (post) if it exists, a link to the next entry if it exists, and a separator between the two if they both exist:

• At the beginning (earliest): next »;
• Somewhere in the middle: « previous || next »;
• At the end (most recent): « previous;

MT provides the mt:EntryPrevious and mt:EntryNext template tags which, in the context of an entry, change the context (think subject of conversation) to the previous or next entry. If that entry does not exist, the content of the tag (should really be element) is ignored.

Simple enough, but how do we implement the separator that should appear only if both previous and next entries exist? Here we go:

What’s going on here? We start an unordered list which is separately styled as a simple inline, non-bulleted list. We enter the EntryPrevious context, which is only evaluated if there is a previous entry — on the very first entry in the archive, this bit is skipped altogether. We then write a simple link to the previous entry (« is the left double-bracket “«”).

Now comes the clever bit: whilst still in the context of the previous entry, we dive into two nested EntryNext elements, which are relative to the EntryPrevious context we were in. These take us back through the current entry to the next one if it exists, where we write out the separator. This bit is only accessed if we’ve successfully found a previous entry and managed to navigate forwards two entries from that.

We finish off by coming out of the two EntryNext contexts, then out of the EntryPrevious context. Finally, we jump back into the EntryNext context to write the link to the next entry, regardless of whether the previous one exists or not.

Examples

I recently needed to transfer some money between bank accounts. I used my bank’s generally-excellent online banking system, and found that they had thoughtfully supplied an example amount. I suppose they thought this example would make it clear that the pounds value goes in the first box and the pennies value in the second, but I think they’ve failed here. I’d be interested to see how many people write the complete amount they want to transfer, including pound symbol and decimal point, into the first box.

The other danger of giving examples is that people might feel bound to use that amount. This may seem ridiculous, but stories abound of people doing really stupid things when interacting with computers, not always in proportion to their general intelligence. Remember, half your users will be below average intelligence (for their population, and yes, I do know about mean, median and modal averages).

Felicitously, the amount I needed to transfer was approximately that given in the example, so in the interests of research, I decided to use exactly that value to see what would happen. Given the unlikelihood of someone intending to transfer precisely that amount, it would have been sensible to provide a warning to prevent a momentary lapse of brainpower triggering an unwanted transfer: you have entered the same amount as in our example — are you sure you want to do this? But no, the transfer went through without warning.

Perhaps I’m being over-cautious and underestimating the general online-banking-using public, but I’d like to know how many transfers of £550.57 there have been in relation to other adjacent figures, and whether the bank staff have a chuckle at their idiot customers any time such a transfer goes through.

Default values

We moved house a couple of months ago. Our first water bill turned up yesterday, showing the meter reading we’d provided when we moved in and the new estimated “reading” on which our bill was based. Now, for a new customer, estimating readings presents a problem as there’s no history to draw on — I don’t believe the infrastructure or processes are in place to allow South Staffordshire to contact Essex and Suffolk to find out how many baths we had per week when we lived in their patch.

So, there are a few things they could have done:

• Ask us for a real meter reading. It’s not the easiest to take, as the meter is buried in the drive underneath a metal cover, a foam frost protector and a thick layer of grime, but it’s not impossible and it’d need reading eventually anyway;
• Use past history from the property, assuming the previous occupant was reasonable with water use;
• Use a default value based on the type of property and type of occupancy.

None of these methods will be correct, but any would be better than the value they chose:

That’s right, they decided that the most appropriate guess for the amount of water used by a family of four for a period of 69 days was none at all.

So our options were to pay the bill consisting of standing charges only, and carry over the water use charges until the next real meter reading, or ring up to provide a true meter reading. Surely it would have been better to explicitly ask us for a meter reading (or even send someone round to take it) rather than send out a bill that appeared not to require any action other than payment, but would eventually end up with a steep increase in charges once the payment deficit became clear?

Summary

These two examples are from seemingly competent companies — the online banking system is great and the water bill is nice and clear — but both examples expose potential error states for users not paying close attention.

When creating examples or choosing default values, think very carefully. Assume a distracted user towards the low end of competence at the task you’re asking them to complete. Do your examples help or generate a risk of incorrect data being entered? Are your default values reasonable? Could it be harmful to the user if they chose (deliberately or through inattention) to use the default value rather than explicitly entering data?

I’d love to hear about other instances of potentially-confusing examples or silly default values: comment away below!

[A/C#] [Am7/G] [Bm7] [Bsus4/E or Emaj9] [D/F#] [E/G#] [Em9]

A/C#

This works for most first inversions (bass note is the third)—just move it up the neck (more on inversions and other slash chords).

   0   1   2   3   4   5    
E  |---|---|---|---|-4-|  A
B  |---|(1)|---|---|---|  C#
G  |---|(1)|---|---|---|  A
D  |---|(1)|---|---|---|  E
A  |---|---|---|-3-|---|  C#
E X|---|---|---|---|---|

Am7/G

This is the second chord in He is the Lord, following on from the standard open G.

   0   1   2   3   4    
E  |---|---|-4-|---|  G
B  |-1-|---|---|---|  C
G O|---|---|---|---|  G
D  |---|-2-|---|---|  E
A X|---|---|-D-|---|  A
E  |---|---|-3-|---|  G

Here, the A string is damped by the ring finger holding the bottom G. The open A could form part of the chord, but the small G-A gap would sound very muddy. This is only really an Am7-based chord in context with the surrounding chords and other instruments.

Bm7

This has a nice full and “open” sound to it. Bit of a stretch for the thumb, but it’s quick and easy to find and avoids barring. Could also be used as D/B.

   0   1   2   3   4    
E  |---|-2-|---|---|  F#
B  |---|---|-3-|---|  D
G  |---|-1-|---|---|  A
D O|---|---|---|---|  D
A  |---|-T-|---|---|  B
E X|---|---|---|---|

Bsus4/E or Emaj9

I use this in place of the E/D# in Let everything that has breath if I’m not using a capo. Some use this as a “cheat” open B, not playing the outer E strings.

   6   7   8   9   10    
E O|---|---|---|---|  E
B O|---|---|---|---|  B
G  |---|-1-|---|---|  D#
D  |---|---|-3-|---|  B
A  |---|---|-2-|---|  F#
E O|---|---|---|---|  E

D/F#

   0   1   2   3   4    
E  |---|-2-|---|---|  F#
B  |---|---|-3-|---|  D
G  |---|-1-|---|---|  A
D O|---|---|---|---|  D
A O|---|---|---|---|  A
E  |---|-T-|---|---|  F#

E/G#

A common but troublesome chord.

My first version is a bit weak compared with the D/F# above due to the missing open A and D strings, and best used as a passing chord only. It can be used for any first inversions up the neck though.

   2   3   4   5   6    
E  |---|-2-|---|---|  G#
B  |---|---|-3-|---|  E
G  |---|-1-|---|---|  B
D X|---|---|---|---|  
A X|---|---|---|---|  
E  |---|-T-|---|---|  G#

Stuart Townend uses the shape below for this chord, but I can’t persuade my fingers to do this. He writes:

On the E/G# subject, the chord I use is 422400, where the A and D string are being barred by my forefinger, the first section of which is flat against the fretboard, but the finger then lifts at the second joint to leave the top 2 strings open! I don’t know if everyone can do this, but that’s how I do it. Sounds quite full.

   0   1   2   3   4    
E O|---|---|---|---|  E
B O|---|---|---|---|  B
G  |---|---|---|-3-|  B
D  |---|(1)|---|---|  E
A  |---|(1)|---|---|  B
E  |---|---|---|-2-|  G#

Eli Awtrey wrote to me with this alternative to Stuart’s, damping the A string rather than using the barre:

   0   1   2   3   4    
E O|---|---|---|---|  E
B O|---|---|---|---|  B
G  |---|---|---|-3-|  B
D  |---|-1-|---|---|  E
A X|---|---|---|-x-|  
E  |---|---|---|-2-|  G#

If you have very long fingers, perhaps you could try this one:

   0   1   2   3   4    
E O|---|---|---|---|  E
B O|---|---|---|---|  B
G  |-1-|---|---|---|  G#
D  |---|-2-|---|---|  E
A  |---|-3-|---|---|  B
E  |---|---|---|-4-|  G#

Em9

   0   1   2   3   4    
E  |---|-3-|---|---|  F#
B O|---|---|---|---|  B
G O|---|---|---|---|  G
D O|---|---|---|---|  D
A  |---|-1-|---|---|  B
E O|---|---|---|---|  E

Sometimes, the lowest note of the chord—the bass note in the music—is not the root note of the chord. These are generally known to guitarists as slash chords due to the way they are notated: (chord)/(bass note). For example, A/G is A major (chord) with a G (note) in the bass. These are usually used in a transitory fashion, to give a sense of progression from one chord to another, such as G-G/B-C.

If you are playing with a bassist or keyboard player who are covering the low notes, you need not concern yourself with the “slash note”: just play the chord on the left-hand side of the slash. If you’re on your own, however, you need to think about emphasizing the bass note to recreate the feel of the music as intended.

There are two special cases of slash chord, where the bass note is the third (major or minor, depending on the chord) or fifth of the scale and appears in the chord. These are called inversions in formal music theory. G/B, for example, is the first inversion of G major with the third (B) in the bass. The second inversion of a chord has the fifth in the bass, such as D/A.

Self test question

What is the second inversion of C# major?

Reveal…

C#/G#. The second inversion has the fifth in the bass, and the fifth of C# is G#.

There are times when it is more correct to write a note as, for example, Gbb rather than F, or Fx instead of G. Let’s look at why — take the scale of A minor (harmonic minor, to be precise):

A-B-C-D-E-F-G#-A

See how every note letter is used in alphabetical order in the scale? OK, let’s look at Ab minor, where every note is one semitone lower than in A minor:

correct:   Ab-Bb-Cb-Db-Eb-Fb-G-Ab
wrong:     Ab-Bb-B -Db-Eb-E -G-Ab

Here, we see a use for Cb and Fb: if we just used B and E, the sequence would be lost. By using Cb, it’s obviously a minor third. Likewise, Fb is quickly identifiable as being the minor sixth rather than the augmented fifth, which would be written as E natural to maintain the alphabetical sequence.

Now think about the diminished chord, which contains root, minor third, diminished fifth and double-flatted seventh. An Abdim chord would therefore be:

Ab-Cb-Ebb-Gbb

The diminished chord is a stack of minor thirds: each note is a minor third (3 semitones) higher than the last. So on a fixed-note instrument like guitar or piano, the same combination of notes make four different chords. Using strict music theory, though, the notes have different names in each chord:

Ab dim: Ab  Cb  Ebb Gbb
 B dim:     B   D   F   Ab
 D dim:         D   F   Ab  Cb
 F dim:             F   Ab  Cb  Ebb

The same principle applies for double sharps. For example, the fifth note in the scale of B major is F#. A B+ chord (B major triad, but with an augmented fifth) would therefore contain Fx rather than G — even though Fx and G are exactly the same note on the guitar and any other fixed-interval instrument.

Taking this to its logical conclusion, it’s possible to construct an argument for triple-sharps and triple-flats. Take, for example, a piece of music in C# major. The major seventh is B#. Within that key, you’d expect the associated simple chord to be B#dim (B#-D#-F#). However, if you use a B#+ chord during modulation, that would contain the notes B#, Dx and… F### (perhaps F#x?). To quote Mary Poppins, that's going a bit too far, don't you think? Comments welcome.

[Posts in this tutorial series]