1) How accurate can we judge the entire planet's average temperature in the year 1800? The graph shows swings from year to year in the 0.2 C range. Can we really judge the average surface temperature of the planet with 0.2 degrees Celsius?
Take a look at the grey band - it's more obvious in the second graph, the 10 year moving average. The grey band is the 95% uncertainty interval for Berkeley's calculation of the average temperature - statistically on each data point there is a 5% chance that the real average temperature lies outside the grey band. You will see that in the year 1800, the grey band is massive: +/- 0.5 degrees. But over time, as there are more measurements around the world, and those measurements have less randomness in them (i.e. get more accurate), the uncertainty shrinks pretty slowly.
2) Also, the chart shows 200 years. This is a blip on the scale of climate science. If you look at the climate history on a much, much larger scale, you'll find that 200 years means nothing. For example, the chart on this page shows that we are much cooler than the average. An sharp increase in average temps would help put us "right".
This is true - no matter how much we heat up the earth, life will survive. But if the climate changes too much from our current conditions, then there will be massive changes. Lots of creatures will become extinct (eventually new ones will evolve, taking advantage of the abundance of food/lack of predators but that happens very slowly) and we will probably have to totally rethink our farming practices. We should move our cities too given that many would no longer be well-situated, but what would probably happen is that we turn up our air-conditioners and burn even more coal. I concede that the effects of climate change are less well understood (at least by me!) than that it is happening.
Or this chart which goes back 4500 years, shows that we just came out of an ice age, so a temperature increase would be expected, and also negates your Berkely graph.
Seriously? I give you the Berkeley graphs, which appear to have used a pretty rigourous method, where you can download their temperature data and source code, and is being peer-reviewed, and you rebut this with a graph that does not have a labelled y-axis and appears to have been drawn with a bezier tool? If you want to convince me that there is no scientific consensus, i.e. that researchers who know what they're doing and are doing it properly, disagree that global warming is happening/is a problem, then please stop using graphs like that. Especially when they disagree with the graph I provided, which gives its sources (IIRC, every temperature measurement they could get their hands on), and includes three other groups' sets of numbers on the same axes - none of which agree with the graph you provided.
Or, finally, this page which shows a whole slew of charts, most of which show that we are in a cold period of climate history, and an increase in average temperature would get the earth back to the "normal" range.
Again, the really-long-range graphs don't have much to do with the current debate, because I'd like life to survive in its current form as much as possible. When large-scale, seemingly-irreversible (on the scale of centuries) changes are made to the only planet we live on, I get nervous about the potential for things to go wrong.
There are too many graphs on that page to go through them individually, but it doesn't give that site any credibility to include graphs like this one, which show very suspicious behaviour - local temperature swings around wildly and then the music stops, when we would expect local temperature always to fluctuate on a year-to-year scale.
Thank you for replying with a little detail. It is good to be able to assess the evidence that others believe to support their beliefs.