No drama, as first worded it irked my experience in multichannel geophysical processing; every data source has potential biases, tweaks, scaling, shifts and normalizing applied to get them to best fit and reveal a larger picture.
In the climate domain the same care is taken to preserve raw data from historic methods, the challenge is syncing data streams that have covariant features (such as matching years against tree patterns from differing logs from a variety of species), in this case logs from a variety of ways of sampling sea water temperature.
Glad you took my comment well, no offense was intended :)
This is an excellent example too -- all too often science reporting will just put in the headline "Updated historical temperatures make threat of global warming (less|more) serious!" -- and folks latch on to "how can history change?!?!" and immediatly go to conspiracy theories, rather than trying to understand the subtlty of what's actually going on
