Bishop Hill More Met Office and statistics
Apr 10, 2013 Climate: Parliament Climate: Statistics Climate: Surface There is more evidence of parliamentarians' attention being focused upon the Met Office's statements climate change. This question to ministers in the Department of Business Information and Skills (the department that sponsors the Met office) has been tabled for a reply next week.
Graham Stringer (Blackley and Broughton): To ask the Secretary of State for Business, Innovation and Skills, whether the claim that (a) every year since 1998 has been significantly warmer than the temperatures you would expect if there was no warming and (b) for the last three decades the rate of temperature increase is significant made by the Met Office in a climate science briefing sent to the Chief Scientific Adviser on 8 February 2010 was supported by any statistical time-series analysis.
The document referred is discussed and can be download here.
Reader Comments (48)
Maybe Graham Stringer should have included this evidence in his question or posed another one.
http://wattsupwiththat.com/2013/04/09/foia-obtained-met-office-document-shows-them-to-be-clueless-about-what-affects-our-climate-and-in-particular-what-caused-the-unusual-weather-last-year/
A must read for all, whether "AGW believer" or "sceptic".
I'm sure that the Secretary of State for Business, Innovation and Skills (who dreams up these titles?) will have a plethora of advisors to provide him with a set of weasel words which don't answer the question (or even tell him he has asked the wrong question and tell him the answer to the question he should have posed).
I reckon Phillip's answered that one!
If AGW was a real thing it should be simple enough to provide a straight answer to a straight question, no?
...was supported by any statistical time-series analysis...
ANY statistical analysis? Of course it can be supported by some kind of statistical analysis. Mann's hockey stick was supported by some kind of statistical analysis. And why hasn't he asked for the details?
So, if I were answering that PQ, I would respond:
"Yes, the two statements you refer to WERE, indeed, supported by statistical analyses. Next question?"
WUWT is now carrying this item - good!
And GWPF!
Very Poor questions ! The one you published recently was th best.
@Dodgy Geezer: the question specifically refers to time series analysis.
if the answer is "yes", then natural follow ups would relate to which particular time series the data was being tested against.
The Met office won't want to be drawn into this sort of discussion as they have no idea what sort of time series best explains natural variability.
I agree with Dodgy Geezer.
The question to elicit the information required, should have asked them to provide the analysis which supported their claim.
Thanks, Matthu - I was with Dodgey Geezer until you explained it.
I would like them to have asked what evidence of AGW they were basing their decisions/arguments on.
At least questions ARE being asked, which is important for the whole world. People everywhere are beginning to get rather annoyed by these faulty claims and the money that's been/being spent. Can't wait for the fireworks to start.
Nice letter in the Times this morning from you Bish. Keep up the pressure!
Apr 10, 2013 at 8:54 AM | Stephen Richards:
I thnk the questions are all coming from the same source. There is a concerted effort to get the Met Office to state publicly what statistical analysis has been used. According to the Met Office statistician, Doug McNeall, they haven't used any, but said they did to be consistent with the IPCC. (Yes, he actually said that. "I believe that the trends were reported as linear as a method of description for consistency with the IPCC Fourth Assessment Report."). But he doesn't believe you can do a time series analysis on the data unless you take into account the physical aspects. He maintains that there has been a scientifically significant increase in temperature, but you can't measure the increase statistically for the reasons given above. Everything he says tells me they've tried to get a statistically significant signal in the data but failed, so have fallen back on the "it can't be statistically analysed" meme. ( I may, of course, be totally wrong, and my impressions are caused by the "communications deficit" widely acknowledged in the scientific community).
Worse yet, I am beginning to think that they don''t have a clue what they mean by scientifically significant increase. When I asked him if he applied the same criteria to the 1880 - 1940 warming did that show scientifically significant warming he told me that was a hypothetical question. It isn't, of course, there should be a set of criteria that they use to define "scientifically significant" increases, and it would be a simple matter to apply them to the 1880 - 1940 period.
I'm left with the impression that the Met Office, or this particular statistician, don't have measurement criteria for "scientific significance", OR they do have them and if they are applied to the 1880 - 1940 warming it is showing "scientifcally significant" warming. Which couldn't, of course, be caused by CO2.
Also, although the government is supposed to answer questions, or explain why they can't answer them, to Parliament, I believe there is a subtle difference between Parliament as represented by the unelected Lords, and Parliament as represented by the elected represenatives of the people, which wwoudl be unthinkable. I am guessing a question there will have more pull than one in the Lords.
Pity the poor statistician in the Met Office as inspired by such as Houghton and Napier. In the outside world, one statistician after another pokes deep holes in the over-inflated balloons of the climatists. Trained to look for substance, trained to be suspicious, immediately, of over-confident claims of evidence and insight, the statistician learns to ask question after question, and slowly clarifies the situation, perhaps with model-building to illustrate points and help encourage further questioning and inquiry. Do that in this new political-goal directed Met Office, and I suspect you'd be out on your ear. Look, the science is settled. We have a communication problem not a statistical one. Why don't you start a creative graphics laboratory and help us out here?
Meanwhile, some doddery retired statisticians while away their days finding new words to old songs. I put a new verse to the Garden Song in yesterday's Met Office post. I was inspired to add another by today's post on Parliamentary Questions by Graham Stringer:
The song is by Dave Mallett and here he is singing it: http://www.youtube.com/watch?v=2m0LewjkO4s
Full lyrics: http://www.sweetslyrics.com/887260.David%20Mallett%20-%20Garden%20Song.html
I think Stringer has got it right.
The more complex you make a question the more chance of getting the answer they want rather than the answer you want.
Dodgy Geezer has, unwittingly, given the right reply!
"Did you use a time-series analysis?"
"Yes".
"What was it?"
OR
"Did you use a time-series-analysis?"
"No."
"Why not?"
Simples.
Never give a civil servant wriggle-room. If you do he'll wriggle.
They think they are very good: http://www.metoffice.gov.uk/about-us/who
"As a world leader in providing weather and climate services, we employ more than 1,700 at 60 locations throughout the world. We are a Trading Fund within the Department for Business Innovation and Skills, operating on a commercial basis under set targets. We are recognised as one of the world's most accurate forecasters, using more than 10 million weather observations a day, an advanced atmospheric model and a high performance supercomputer to create 3,000 tailored forecasts and briefings a day. These are delivered to a huge range of customers from the Government, to businesses, the general public, armed forces, and other organisations."
"Climate change has become an increasingly important issue and our research continues to create an ever clearer picture of how it will affect the planet and our lives. This plays a vital role in providing evidence to support climate predictions which show the planet is now locked into at least 2 °C of warming and rapid reductions in greenhouse gas emissions are required to ensure this does not rise further for future generations."
Agreed. I would also like Stringer to ask the decision-makers how many years of non-global warming we must endure before they accept that CO2 is evidently a very minor factor in determining average global temperatures, and that Climate change science isn't settled.
I see some interesting discussions on the utility of PQs.
I am not sure that anything useful will come from a PQ, or, indeed, has ever come from one. I base this belief on Darwin's Theory of Natural Selection.
PQs have been around for a long time. If they were useful to the ruled, they must, perforce, be damaging to the rulers (AKA Civil Servants). If they were damaging to Civil Servants, then the Civil Servants would have taken steps to abolish them. PQs still exist. Therefore, they cannot be damaging to Civil Servants, and hence of use to us.
The above argument is a bit tongue-in-cheek, but I believe that it contains some truth...
I am with Dodgy Geezer on this one These actions can do no harm...but do not expect anything useful or meaningful to flow from them.
Think of the question as a small ball-bearing and Parliament as a great big sloppy super-saturated sponge.
Having said that...Graham Stringer is certainly starting to "get it". Hurrah.
All this stuff about time series analysis seems to presume that there is an underlying (albeit unknown) well defined stationary process generating the time series representing (say) global average temperature.
Years back, communications network engineers made similar assumptions about datacommunications traffic. It slowly became apparent that the process generating the time series, far from being well defined and stationary, was in fact nonstationary. Even worse, its nonstationary characteristics varied with time in an unfathomable random fashion.
I am all in favour of asking the Met Office difficult-to-answer questions about their use of time series analysis underlying their statements of things being "statistically significant".
But, in reality, there seems good reason to believe that the climate system is a complicated chaotic oscillatory system, switching between modes in unpredictable ways. To represent such a system as a stationary autoregressive/moving average system driven by stationary white noise seems to me a gross oversimplification of the reality.
I'm starting to believe that it is inherently unmodellable and that questions about "statistical significance" of observed climate observations are inherently unanswerable, even by persons whose first name is "Doug".
Martin A. Agreed, we are a long way off knowing enough to model the climate and since it's essentially chaotic will never be able to use the models to foretell future states.
However, having said that, there is evidence that the Met Office is hiding the fact that the temperature increase of 0.8C isn't statistically significant. So whether it can be modelled, or whether the time series has any relevance pales into insignificance beside the fact that a scientific organisation, for which we pay £200 million/annum deliberately hiding the facts (as they see them) from the public to push a political agenda. See below a note from Julia Slingo to John Beddington dated Februaruy 2010.
(4)I have also asked my staff to prepare a summary document on our recent work with releasing surface temperature observations and the code, along with showing that the global warming signal is robust. I expect to send that on to you later this week. Hopefully that will be enough to convince Muir Russell not to call in statisticians to repeat the exercise. We would of course be prepared to brief him on our findings .
Note she doesn't want external statisticians looking at their work. Why would that be do you think?
Don't underestimate Graham Stringer. He's a shrewd and effective politician. He knows what he's doing.
And like many of the good guys, he trained as a chemist.
The Bish himself, Jonathan Jones, Ruth Dixon, modesty forbids...........
and many others who have their scientific feet firmly in experiment and observation, not in theory and models and belief/faith.
geromino,
The Met Office was (still is, to some extent) an organisation that attempts to forecast the weather.
If you think it has also become a scientific organisation, you are mistaken.
It morphed into a propaganda generator (with weather forecasting as a sideline), not a "scientific organisation".
If the facts are convenient, it will trumpet them (“in the last 130 years, ten of the warmest years on record have all occurred since 1998″). If the facts are inconvenient, it will do all it can to hide or minimise them.
Prof. Slingo really is a piece of work.
'Hopefully that will be enough to convince Muir Russell not to call in statisticians to repeat the exercise'
does not fill me with confidence that she is committed to full and open disclosure of her teams publicly-funded work.
I was also surprised that her published CV does not seem to list her academic qualifications. Normally professors are delighted to tell you how clever they are.....
http://www.ugamp.nerc.ac.uk/cgam-trop/vita1_www.html
Martin A at 11:14 AM notes 'All this stuff about time series analysis seems to presume that there is an underlying (albeit unknown) well defined stationary process generating the time series representing (say) global average temperature.' and goes on to wonder if ARIMA models are a gross oversimplification.
I share his concern in general, but I would note that the climate system has influences acting on a wide range of timescales, and that perhaps if we look at this range in chunks we can do some worthwhile time-series modelling analysis as and when data can be found? Generally such modelling assumes that there is an underlying, fixed system of causes and this may be approximately true under some circumstance. For example, over periods of a hundred years or so, we might regard some orbital parameters as constant, and over months we might regard some solar ones as constant also. For known cycles or trends, there are options for including, sometimes by extracting (!), things we know or presume e.g. decadal-level cycles during the 20th century in some statistic or other, such as global mean temperature. There will be regime changes that in due course will undermine such models, ie. when it no longer becomes tenable to talk of relevant influences as being unchanged.
You can see this in monitoring processes on the factory floor, when longish periods of stable behaviour are disrupted. Perhaps a supplier of raw materials has made a specification change without telling you, or the night shift have taken to doing things their own way.
In a nutshell, I would not discount the value and relevance of time series analysis of climate records. It just needs calm and careful minds to conduct it - requirements which seem to exclude alarmed ones.
@ Joe Public
"...I agree with Dodgy Geezer...."
I like this interchange. Particularly with these screen names. Sort of sums up the climate change fiasco, indeed, the whole human condition...
geronimo
Note she doesn't want external statisticians looking at their work. Why would that be do you think?
*************************
Standard operating procedure.
Think of the Magic Circle
Motto - Indocilis privata loqui
"The history of magic as an art form has been shrouded in mystery in order to protect secrets, often being handed down verbally and to only a select few,......"
http://www.themagiccircle.co.uk/about-the-club/ourmuseum
Martin A
Isn't all that what the IPCC itself admitted?
That climate is a chaotic system and trying to forecast what it was going to be half-a-century down the line was not possible.
So why are we paying attention to the rubbish that is being churned out now rather than the more cautious and realistic approach then?
{Answer not required, by the way :-)}
Met Office Mandarin1: "Slingo says some awkward MP is asking questions about whether the recent warming is statistically significant".
Met Office Mandarin2: "No problem, to save having to get some one in to actually do a statistical time series analysis, just look at these historic temperature datasets, and you can clearly see the late 20th century uptick:" - NikfromNYC's graph - http://oi52.tinypic.com/2agnous.jpg
Met Office Mandarin1: "Okay, er, eh, oh... wait a minute... are you sure?"
Met Office Mandarin2: "Oh dear, we really did fall for Napier's green groupthink bollocks didn't we...".
Martin A
"If the facts are convenient, it will trumpet them (“in the last 130 years, ten of the warmest years on record have all occurred since 1998″).'
I assume this is taken from 'the lovely' Kate Willett's non-explanatory video entitled 'Where's the proof the earth is warming' which can be found here - http://www.myclimateandme.com/2013/04/09/new-research-suggests-transatlantic-flights-to-get-more-turbulent/#comments - I don't know what the relationship is between the Met Office and mycliamteandme but I see the MO logo proudly displayed in the header. Anyway, as far as I can see the main purpose of myclimateandme is to ramp up the alarmism. A few more enquiring minds apart from yours and mine would make it more interesting.
I read elsewhere that Prof Slingo's qualifications include a BSc in Physics from Bristol.
Perhaps she was just short of space on her CV.
We already know that at 20th century warming rates it takes 15-20 years of data to demonstrate a 95% significant warming. Phil Jones said as much in his interview.
What concerns me is that Stringer and others would like to claim that warming has stopped on the basis of much shorter periods of data. Statistically all they can say is that in the 21st century they are 95% confident that the temperature has increased by no more than 0.2C or decreased by no more than 0.2C.
It is ironic to see them attempting to use statistics in an attempt to discredit previous warming and then ignoring the statistical weakness of their own arguments.
EM I was convinced for a long time that you were being deliberately obtuse when you posted here, but now I'm not so sure.
Stringer doesn't want to say the temperature at the end of the 20th century wasn't high, it was we measured it with thermometers. The point is that the Met Office described it as a significant rise in temperature. Notwithstanding the discussion going on between Martin A and John Shade, which I won't ask you to read, there is a time series of temperature records from 1880 until 2000 from which the Met Office describe the rise in temperature as "significant". In real life world outside the Looking Glass to prove significance you would build a statistical model against which you'd test the data for significance. The Met Office claimed they had, they also claimed they hadn't, (I'll bet Lewis Carroll is kicking himself for dying so soon).
What, you may ask, is all the fuss about "significance"? Well if the rise in temperature doesn't have statistical significance then it cannot be assigned to the human emissions of CO2 because it's within the natural variations of the time series, and the Met Office has been telling porkies to our politicians.
It is highly unlikely they wouldn't have tested the series with a statistical model, but the one they claimed to test it with is hotly disputed as the correct model, and, of course doesn't provide a statistical significance to the 0.8C rise in temperature between 1889 and 2000. Hence Dr Slingo's worry that Muir Russell would ask professional statisticians to look at the Met Office and CRU temperature data records.
A chap called Doug McNeall, a statistician at the Met Office, has now popped up and told Doug Keiller that the time series cannot be tested for statistical significance, but there is another "significance" called scientific significance, so the 0.8C rise has scientific significance. I'd never heard of such a construct and asked how they got scientific significance, in an exchange at Doug's site. it was all very polite, but he's now told me he's too busy to answer questions, which he may be, but if it exists he could simply have pointed to it in the literature.
So nothing to do with lowering temperatures. And if you want a professional explanation without the mistakes I've put in I suggest you read Doug Keiller's.
Graham Stringer is emerging as one of the heroes of the politico-scientific struggle to impose a proper perspective on the claims of climate scientists. Doug Keenan's work is beginning to get traction.
I am rather hoping Stringer's supplementary will be along the lines of "How much would the British taxpayers save on an annual basis if the Met Office shut down its climate research activities with immediate effect and concentrated on weather forecasts?"
http://www.metoffice.gov.uk/research/monitoring/climate/surface-temperature
From the Hadcrut 4 data you can answer Mr Stringer's questions for yourself using a simple statistical test. Compare the 95% confidence ranges for any two years of your choice. If the two ranges do not overlap, then you can be 95% confident that one is significantly warmer than the other.
Question a) is meaningless without a reference date. One possible answer would be "Yes, every year since 1998 has been significantly warmer than 1992." If you wanted a No, you could choose a relatively warm date in the 1990s. If you go back to 1970 for your reference date the answer would be an unequivocal Yes. If you use the document's 1861-1900 baseline the answer would be even more emphatic.
Question b) is a Yes.
Consider 2011 with a temperature anomaly of 0.40 and a 95% confidence range of 0.31 to 0.50.
Compare it with 1987 with a temperature anomaly 0.19 and a 95% confidence range of 0.10 to 0.27.
The upper bound of the1987 range does not overlap with the lower bound of the 2011 range. 2011 is significantly warmer.
Essentially , of course, the question is political instead of scientific. Whatever the answer, Mr Stringer will use it as an excuse for a sceptic speech.
After watching This Morning With Phill And Holly then Loose Women with Janet Street Porter i tucked into ITV Lunch Time News
Lawrance McKinnty ITN Science Corrrespondant was live form the Arctic Circle .So he started is the recent extended Cold Snap and Wettest summer in The UK in the 100 years the result of Ice Melt in the Arctic caused by Global Warming?.
Lawrence NO
What was the wettest summer in the last 100, 200 ,300 ,400 years then Lawrence
The 17 Century flood that washed away the original Tyne Bridge.Rebuilt it with Iron.Still there today.
Ice melts in the Arctic and then returns every year with the season and the Planet orbits around the sun Lawrence.
And the recent cold snap . I wasnt around in 1947 but i heard that was really bad hasnt been one like it since the 18 century Regular occurance.Read the Little Match Girl by Charles Dickens Lawrence.Sad ending she died in poverty on a London street from Hypothermia. What inspired Dr Barnardos Lawrence.
So Lawrence what happened to the snow they said we were never going to see again.That homeless girl today would be alive now.Moving the Goal Posts changing your tune no more Scorched Earth Policy then Lawrence.Its Weather Weirding now Lawrence .How convieniant Lawrence.Hotter Summers and Colder Winters really Lawrence.Hedging your bets then Lawrence.We shall see Lawrence.
There is nothing unusual about the Weather.The Arctic and Antartic are still cold come home Lawrence.
"If the facts are convenient, it will trumpet them (“in the last 130 years, ten of the warmest years on record have all occurred since 1998″).'
I assume this is taken from 'the lovely' Kate Willett's non-explanatory video entitled 'Where's the proof the earth is warming' which can be found here - http://www.myclimateandme.com/2013/04/09/new-research-suggests-transatlantic-flights-to-get-more-turbulent/#comments - I don't know what the relationship is between the Met Office and mycliamteandme but I see the MO logo proudly displayed in the header. Anyway, as far as I can see the main purpose of myclimateandme is to ramp up the alarmism. A few more enquiring minds apart from yours and mine would make it more interesting.
Apr 10, 2013 at 1:48 PM nTropywins
I think it is a favourite saying of Slingo's too.
My Climate and Me is a Met Office website pure and simple. Obviously meant ramp up alarmism, as you rightly say.
I share his concern in general, but I would note that the climate system has influences acting on a wide range of timescales, and that perhaps if we look at this range in chunks we can do some worthwhile time-series modelling analysis as and when data can be found? Generally such modelling assumes that there is an underlying, fixed system of causes and this may be approximately true under some circumstance. For example, over periods of a hundred years or so, we might regard some orbital parameters as constant, and over months we might regard some solar ones as constant also. For known cycles or trends, there are options for including, sometimes by extracting (!), things we know or presume e.g. decadal-level cycles during the 20th century in some statistic or other, such as global mean temperature. There will be regime changes that in due course will undermine such models, ie. when it no longer becomes tenable to talk of relevant influences as being unchanged.
You can see this in monitoring processes on the factory floor, when longish periods of stable behaviour are disrupted. Perhaps a supplier of raw materials has made a specification change without telling you, or the night shift have taken to doing things their own way.
In a nutshell, I would not discount the value and relevance of time series analysis of climate records. It just needs calm and careful minds to conduct it - requirements which seem to exclude alarmed ones.
Apr 10, 2013 at 12:47 PM John Shade
John,
I'd agree with you if we had detailed records going back 100's of thousands of years. Tests for stationarity could be applied to sections of the records. Dynamic models could be constructed from the observed behaviour and then used to detect sudden deviations from stationarity, as you suggest.
But in reality, with records going back a few hundred years, at best, it is as if we had a couple of hours of recording from your factory floor example. Simply not enough data to do any useful analysis with.
I'll stick with my unmodellable, unknowable hypothesis for the climate system. But always ready to change my opinion if someone really did come up with verifiable models.
That's a powerful point, Martin A. Prof Frank Ludlam, would that he had lived to have helped squash the climate hyperbole, told we students back in the 1970s that climate data was so sparse, the analysis and prediction problem was a bit like trying to forecast the progress of a football game given a random snapshot of part of the field after kick-off, and given a poor grasp of the rules. Ah but we have computer models, I can hear the newcomers cry. I think Frank would have warned them that their models might get in the way of their thinking about the real system.
But coming back to the data, the observations and the reconstructions. On the scales of the climate system, we don't have very much, and there are problems with all of it. On the scales of our minds and knowledge, we have great piles of the stuff. We can hardly stop ourselves poking it around and seeing what might be learned.
So I'll cling to hopes of others squeezing some value out of statistical analyses yet, even, perhaps especially, if they clarify just how little we know and how far we have to go before a shred of clothing might be found for the Emperors of Alarm.
Regarding Prof Slingo's qualifications, there is a biography on the Bristol University website from when she was given her honourary Doctor of Science a couple of years ago.
She appears to have a BSc in physics and maths from Bristol from 1973, and then a PhD described rather oddly as follows:
"In 1986 Julia moved to the National Centre for Atmospheric Research in the US. It was while she was in the US that she completed her PhD, externally, in the Physics Department at Bristol University. The topic was atmospheric physics, and it was a thesis completed through a series of published papers."
Other links suggest her PhD was actually awarded in 1989. Looks as though there is no stand-alone thesis or singular topic, but that she was awarded it on the basis of continued research. Have to say I'm not sure of the precedent for this in the UK - not something I've heard of before.
Way back- in the 1970s and 80s - senior civil servants I came across were seriously concerned about questions to Ministers for fear of getting them into trouble by providing incorrect answers. Of course back in 2004/05 it was the explicit policy of government to ignore the "deniers" (cf Brown`s remarks about "flat earthers"). Such a response will not be so easy today.
No doubt Mr Stringer is awaiting the reply with interest. No doubt he will have his supplementary questions lined up according to the answers forthcoming from Mr Cable or whoever deputises. An attempted cop out would be a written answer but that would only buy time.
A more devious thought is that the question has been set up between Mr Stringer and someone in the Department in order to pressurise the Met Office following Lord Donaghue`s questions in the Lords (to which no satisfactory rely has yet been given). Now who recently moved to theDept for BI & S? This merely my speculation.
Ian Blanchard,
Getting a PhD by publishing a sequence of papers is certainly an option which has been taken up since Slingo's award. Plenty of advantages to both the candidate (and institution) this way, not least a publication record and a pretty strong position at viva having been peer-reviewed several times already. Haven't come across a situation where papers haven't been collated as a broader thesis though.
Hi Latimer. My BSc was Applied Chemistry.. My MSc IS Engeneering, Cybernetics (computer modelling, in the same lab that had Loverlock as a visiting prof)
Barry Woods
I see you have been invited to meet Richard Betts and that the meeting is likely to be vidoed and aired on myclimateandme. Can I suggest you take your own film crew with you to make sure the version offered by MCAM is based on observations rater than models.
@Barry
Thanks for the reminder. I remembered from our last pub meet with Rhoda (computing) and Jonathan (chemist) and Ruth (chemist) and our 'Lurking M's (chemists) that we you and I shared experience of computing...but I wasn't sure enough of my memory that we were both chemically trained to include you in my list
No offence meant...just my poor recollection.
All other chemists are invited to declare themselves.......
PS: Be gentle with Richard. He finds himself in a tough position........
Can't wait to meet the Met Officer's Brand Manager...... and hopefully get an answer to a 2 month old question I set the Chief Press Officer..
and of course have a laugh at Lewandowsky... I might take a copy of all the academic misconduct complaints along, for those in the 'office' to have a laugh at.
Barry Woods
you have come a long way since you mis-spelled your way through all those awful Richard Black blogs when we were all in short trousers. A film star already! (well almost) enjoy