The end of an era, or should that be epoch?

Yesterday was the last day of the move of the University of Oxford Department of Earth Sciences from the building(s) it’s occupied since 1948 to the new building just around the corner. (The chemistry labs are still situated there until November but the building is now technically “owned” by Chemistry.)

This was the end of an era for me as well. I was the first person to move into the extension in late 1992, getting key number 139, and my office was one of the last to get moved out as well. So, I’ve certainly been the longest resident. Yesterday I handed that key back for the very last time, after having it in my pocket continuously for 18 years almost to the day. The end of another era.

The new building is very different from the old one. It’s light and airy but also more “corporate” and soulless. One real benefit is, however, that the common room is on the roof and last night, at the weekly Happy Hour, it was very pleasant sitting on a sofa with only the spot lights switched on watching the nearly full Moon climb above Headington Hill intermittently shrouded by scudding, silver-rimmed clouds and the sky changing from a light turquoise to a dusty dark blue over the period of an hour or so.

Still, the only constant in life is change and I can see that the new building *IS* an improvement in many ways from the old one. There are things the old building had which were better, such as a larger library, but on the whole there are fewer of those than than the new one’s advantages. Let’s just see how things progress…


In a spin: Speed cameras and the fallacy of road safety.

Recently, with the cut in funding and subsequent turning off of speed cameras in Oxfordshire there has been a great deal of chatter about how this will increase the numbers of deaths on the roads. Today had the local speed camera organisation, “Thames Valley’s Safer Road Partnership”, spouting statistically unsustainable statistics about how many people were now breaking the speed limit as opposed to when the cameras were operational. (Notably, they didn’t say by how much the speed limit was being broken by. Also, one location is notorious for chavish lads driving recklessly, which wouldn’t be halted by the camera.)

Two days ago we had another pressure group and a number of police chief constables put their oar in: One basically saying that there would be carmogedden and the other blaming the middle classes for speeding “until one of their family got killed.” That second one is laughable if you’ve ever lived anywhere close to a “working class” neighbourhood.

The trouble with all this spin and hoo-hah is that the evidence for speed being the main killer on the roads is just not there. In 2008 only 14% of road accidents had speed as a contributory factor. Note that this is ALL accidents reported to the police and not just lethal ones. So, speed cameras, even if they were 100% effective, would not help in 86% of accidents.

So, now we know that speed is not as high a risk as the quangos who run the cameras and the police forces (who don’t want to spend money on traffic enforcement officers) would make out, do cameras actually help at all?

Well, in some locations I’m sure that they do. In most locations, however, they’re more of a hazard than a help. How many times have you seen a car brake heavily (usually from a speed under the speed limit already) to pass a speed camera? I’ve seen it quite a number of times, causing the cars behind to have to make sudden braking maneuvers themselves. i.e. increasing the risk of an accident. Also, forcing drivers to be more concerned about the absolute speed of their vehicle and watching the speedo more and more decreases the time they have to concentrate on other road hazards or taking note of what an appropriate speed for the situation might be, again increasing the risk of an accident.

If you add to this the congestive effect of cars slowing to well below the speed limit to pass the cameras and, on a road close to capacity, causing the generation of a solitary wave of congestion to pass back along the queue causing a complete jam.

What about the other “safety” features imposed by local authorities on the roads in the name of safety?

Well, there are the extensive 50 mph speed limits which seem to encompass huge swathes of the A road network. It seems that the default reaction when there are an above average number of accidents on a road is to restrict the speed limit by a mere 10 mph. The fallacy here is that doing so will automatically cut incidents.

Let’s break down the logic using a specific example, the A420 between Oxford and Swindon. There are a few places where the junctions are relatively blind and would benefit from the traffic being slowed. It could be said that in these areas 50 mph is still way too high and a 40 mph limit would be more appropriate in the close vicinity of the junction. However, for most of the road the national speed limit of 60 mph is not a dangerous speed (under good road conditions). The majority of fatal accidents on the road, however, are from drivers who have been reckless and have been speeding, losing control of their vehicle on bends. Note: these people are already breaking the speed limit.

So, the knee-jerk reaction has been to decrease the speed limit on about two-thirds of the whole length of road to 50 mph. Firstly, his will not help to greatly reduce the danger on the risky junctions as the time to see a vehicle from the junction before turning out onto the road has not decreased by much. Secondly, those who caused accidents by speeding recklessly will still ignore the speed limit. (And no, more speed cameras wouldn’t help.)

In other words, the responsible majority have been hugely inconvenienced and no real good has come from it other than to allow the councils to say that they’re doing something.

Of course, I am talking heresy. Tut-tut. The slogan says, “Speed kills!” but actually it doesn’t. Changing the slogan to “Inappropriate speed is highly risky” wouldn’t go down well as it’s not punchy enough. Of course, it’s not the speed which kills ultimately, it’s the rapid deceleration.

Just my two-penny’s worth…

And the Bank Holiday weekend flu by.

Well, this last weekend was a Bank Holiday and my parents had come up from Cornwall to see me. Unfortunately I managed to pick up a rather nasty flu virus which started to show itself the evening they arrived, last Thursday.

Initially I merely thought it a minor cough and it wasn’t really until Saturday morning that it really hit me. From then on, until today really, I spent most of my time in bed. Not a great way to entertain my Mum and Dad. Not only this but I’ve also managed to pass on the infection to my Dad.

So, a Bank Holiday to forget really.

Summer here?

Well, it seems as if Summer has arrived early. However, if the last three years are anything to go by, May will be the warmest month and so this IS the Summer. Make the most of it!

The blog site is now configured.

Well, after some advice from Alec Muffett I’ve got WordPress configured in a manner which is reasonable, so I can release it to the world.

I must admit that adding a WYSIWYG editor for posting has made the whole thing a far more pleasant experience.

There are a few things I thing I’ll change but on the whole the simple, minimalist theme I’ve chosen fits in well with the minimalist home page I have.

First Post

“And so it begins.” Immortal words.

I’ve set this blog up mostly so as to document my up-coming sponsored kilt wearing event.

First of all, I’ve got to work out a number of details such as modifying WordPress so as to make it look presentable. I’ve also got to order four kilts and the accessories such as a pin or two, sporan, sporan chain, belt etc.

Building your own PC is now too costly: Why self-build anymore?

Over the last few months I’ve been investigating the replacement of my current games PC. The guts of the system are now approximately five years old and have had a mid-life upgrade of processor and graphics card to the fastest that the Socket 939 and AGP ports can support. It’s time to move on.

Wanting to make any new system last as long as the old one I’ve been waiting for the latest generation of Intel chips to become widely available and the price to drop below their launch price, and this has now happened. So, I’ve been researching the components in detail.

It is quite shocking how the price of all the self-build components, except hard disks, has shot up over the last few years. In 2005 it was quite easy to pick up a high-end motherboard for around £80 and the processor to go with it for about £100. Graphics cards of the generation before the current "greatest thing" were again around the £100 mark. Today we find this somewhat different with the motherboards four times the price, the processors well over doubled, as have the graphics cards.

This morning I finally spec’ed out the PC components for a new machine: Gigabyte X58 Extreme motherboard, 3GB DDR3 memory, ATI 4870 graphics card, power supply, etc. and it came out to ~£1500. More if I wanted to go for a SAS boot disk. All this with no warranty and I’d still have to add the Windows tax on top.

Just a moment though. ~£1500 is the educational price of the entry-level Apple Mac Pro! This has a Xeon verson of the Core i7 chip I’d spec’ed, so it has a far bigger cache. The system also has ECC RAM. It just doesn’t have the same ATI graphics card… but it can for ~£130 and for a further £50 it can have 3 years of on-site maintenance as well (and it comes with a decent UNIX instaled as well). Why should I go to the hassle of building my own system anymore?

The upshot of this has been that my new Windows games/photographic processing box is going to be a Mac Pro. I’ve spent a little extra and gone for the hghest spec. processor so that makes it a little more future-proof.

[Retro computing] Time passes, processors go faster, basic tasks take the same time.

This morning, having been playing with the Sinclair QL last night and having it still sitting on a coffee table I decided to see just how different the time it takes from powering up a computer until it’s ready for a basic task has changed in the 25 years between the QL’s manufacture and my MacBook Pro.

So, I devised a race. Which one could boot up from cold and load a spreadsheet application and be ready for me to start inputting a few numbers to add up. A simple task you might want to do any day and, once the machine is booted, mostly an input speed limited one rather than that requiring processing power.

In the red lane we have: Sinclair QL, 7.5MHz 68008, 640KB RAM (Sandy SuperQboard expansion), QL Abacus on Microdrive.
In the blue lane it’s: Apple MacBook Pro, 2.6GHz Intel Core 2 Duo, 4GB RAM, Numbers (iWork’08) on hard disk (7200rpm).

And they’re off….

Sinclair QL gets to the Abacus input prompt in 1’9"
Apple MacBook Pro get to the Numbers input prompt in 1’7"

I know that the two spreadsheet programs are vastly different, but for the task I’ve outlined they are both capable of doing the same thing equally easily. So, in 25 years there’s been 2 seconds taken off the time. But remember, the QL was using a slow tape drive to load the program, it would have been about 30 seconds faster if it had been on floppy disk.

Progress s a wonderful thing! 🙂

I’ve been rather neglecting LJ recently. I’ve nothing useful to say really as life is merely trundling on.

Christmas and New Year went by O.K.

I bought my Dad a Meade LX90 8" telescope but we didn’t have the weather to play with it a great deal. It does have a minor fault, the positive power lead from the battery compartment’s disconnected, but seeing as it’s going to be powered externally anyway it’s not a problem. The only problem was that the software and communication cable didn’t arrive in time for me to take them down to Cornwall with me. So, instead of my parents coming up to me for my birthday I’m going down there, along with all the technical gubbins.

Before Christmas I managed to rescue from the Physical Chemistry Department at work a bunch of old Sinclair computer equipment. Basically two ZX Spectrums and a QL. I found that one of the Spectrums (the newer, Spanish built one) was dead but the other was very much alive (but with a duff keyboard membrane) so I managed to build one fully operational Speccy. I couldn’t test the QL then as it didn’t have a power supply with it.

This re-ignited my retro-computing interest so i brought my old speccy rubber keyed case back with me from Cornwall so that I can re-fit the circuit board from the speccy+ "upgrade" case it’s currently in. I’ve just received a new keyboard membrane for it and also a QL power supply for the QL I obtained.

The QL, it turns out is a strange one… one of the ROMs is missing so that on its own it only half boots, From the socket tarnish it looks like the ROM’s been gone for many a year. Wierdly, however, if I pug in the "Metacomco QL PASCAL" external ROM (which came with this machine) it’s happy and boots normally.

I also had to fix the keyboard connector. As usual one had cracked, but by trimming the end got it functional again. I was even able to test the microdrives as it came with the Psion office suite. They’re still working after all this time. I can remember when the QL came out that the pundits couldn’t wait to damn the microdrives, deeming them unreliable. How wrong they were.

Courier madness

I’ve ordered some parts for my Dad’s PC down here in west Cornwall and they’re due for delivery today. So, I decided to take a look at where Citylink’s “local” offices were so as to gauge when the delivery window might be.

Can you believe that Citylink’s “local” depot is actually in Okehampton, Devon?! How on earth can they afford to do that?

For those who don’t know the geography of the area, Cornwall from tip to the border is approximately 70 miles and Okehampton isn’t on the border. So, for a deivery van to deliver to west Cornwall it’s a 200 mile round trip. Now, if you take into account the travel times for a van of about 2 hours from depot to the start of the delivery cycle and add haf an hour or so at the end of each day for the driver to do the paper work and assume that each drop-off of a parcel i going to take half an hour to allow for travel time, that means that there can only be 6 parcels delivered per van per shift. Seeing as the fuel for merely the round-trip (without the delivery)  would probably be in the region of 6 gallons of DERV, this means that each parcel will have cost the delivery firm approximately £7 in fuel alone just for the tertiary part of the journey. This is already the total purchase price of the courier service.

Surely, given current fuel prices, this is madness and it would cost the company a whole lot less to have a more local delivery office, as they used to? After all, having the nearest office a quarter of the way to London they might as well go the whole hog and deliver to the whole of the southern half of England from one depot in Swindon and be done with it!