I suspect they rebuilt from a manual copy on someone’s desktop months or even years old.

True redundancy and live failover would have been overkill in this situation. It would be normal to have to build a new server from day/week-old backup tapes, which might take a day or two – that wouldn’t be terrible in this context, although it would be catastrophic for, say, a bank.

he’s no dummy.

There is evidence that this was not a fault tolerant system with live backups in place or any failover. See the screen grab in my post noted above. It is likely they had a single old windows server performing multiple functions with poor if any security between the trusted network and the Internet. Their “emergency” server was probably something that once or twice a year someone manually copied some files onto if that frequently.

Within an hour or so of the inquiry being announced I emailed asking if it was possible for a Canadian to be invited to testify and, if so, advising the committee to make you the top priority. They didn’t reply and they didn’t comply. Instead they invited Lawson and Peiser (30 mins) to go up against Acton, Jones, Beddington, Slingo and Watson (at least 70).

It’s one thing to think that you’d be the best person to shed light, quite another to identify with you as you listened to the resulting session on the first day of the month and read the final report on the last.

One thing I think we all had confirmed is that anything out in the open, however inadequate, has tremendous value, as became clear as Quentin Letts and the other parliamentary sketchwriters got their teeth into it the next day. (Quentin, who I now see was voted political journalist of the year by his peers last year, only just received my email appreciation and replied very nicely. I’d forgotten my younger brother had stayed with his family in the 70s and all that kind of thing. Anyhow.) The sketchwriters quickly picked up that the questioning had been soft and the answers unimpressive, with hardly any technical background. Just imagine what they’d have said with tough questioning, Steve and Ross opening the batting and equal time overall between heretic and orthodox. But the glass half full – or even a fraction full – was better than none, because it was done in the full glare of normal parliamentary reporting. That’s why the committee’s recommendation that Muir Russell do as much as possible in public is also spot on. (Not that I expect Russell to comply. But then I didn’t expect Climategate. We can deal with the situation either way, given the blogosphere.)

Another important observation is that the committee’s report had far less impact in the media in London than the oral evidence, in my estimation. People, including sketchwriters, prefer to make up their own minds. The blogs no doubt increasingly help with that. Well worth noting (and why I don’t really see it as George M does).

It’s been entirely right for you to make clear how inadequate the inquiry has been in the areas you are most familiar with. It was also understandable that you should initially blame Slingo for her email response – and notable that blog critics on that specific point were a help. The real issue on reflection is Briffa’s work on that section of AR4. I believe you.

Nigel Lawson is another thing. What does one say? If you’d had 40 mins, like Jones, then 30 for Lawson later on wouldn’t have been that bad an idea, in my view, as I’m sure he would have deferred to you on the science. Lawson’s area is climate policy, outside the strict scope of the committee. There again, they felt free to go outside scope in questioning others. Who knows. Lawson’s an important British parliamentarian and no doubt that’s what counted. I’m grateful for the stance he’s taken on AGW, which has been a lonely one. I don’t want to diss him that much. The fault was with those who set up the oral evidence.

Anyhow, it makes sense to me that Briffa’s ‘trick section’ on Wahl, Ammann and Wegner is the real problem and that it’ll take you a little while longer to parse, to explain in detail why. We’re massively in your debt that you took this area up with that email to Mann eight years ago and haven’t let go. You could only ever attack it in a way that made sense to you. We’re all ears.

Just to clear things up, I should add that it is entirely probable that the police did take the server in evidence. Speaking as a sysadmin, it’s a basic principle that servers should be presumed to be likely to fail. The loss of one server shouldn’t cripple anything. In a setup where you have more than one server, it really shouldn’t even result in a hiccup in the service received by the users. Hardware is cheap, and should have redundancy/failovers. Assuming a near-total lack of forward planning, the worst that should happen is a brief outage whilst a new server is express-shipped and has the most recent backup written to its hard-disk.

(P.S. If your company experiences worse than this, let me give you my number…

I just tune them out – they occupy the same space in my brain as the other screen fixtures like the title bar. No difference to me, but the ad company still pays up.

Once secured the drives can be removed and mirrored onto others for return to the owner.