Forget the East Anglia CRU e-mails, folks..

A lot of attention is spent dealing with the "hide the decline" and other conspiratorial comments and an equal amount of energy is expended by the desperate defenders of AGW inventing tortured explanations for their heroes at the East Anglia CRU. But what is being overlooked is something that is even more damning than the memos and e-mails: it's the actual source code of the programs that are responsible in part for the analysis and results these clowns have foisted off on the public in an attempt to Hoover up grant money.

My background is that of a software engineer - a computer programmer - and I have been doing this for over forty years. I have worked as a consultant for 20 of those years, written a quarter of century of COBOL programs (and probably 20% of all of the Y2K bugs), written C, C++, Java, Jovial, Fortran (ugh!), Basic (MS and Visual), and even RPG. (Although, to be fair, someone once said that there was no such thing as an RPG programmer, only coders.)

[..further gratuitous, self-serving recitation of personal experience redacted..]

I have written my own software and maintained others' work. Some of the stuff I have worked on was really good and some of the stuff was 100% grade A crap. It looked like the person who wrote it was taking drugs or drinking heavily.

Anyway, enough of the CV horn-blowing. Where I am going with this is that I have looked at some of the source code files released in the CRU hack and, at best, they fall into the latter category of garbage.

In the clip above, what the guy is talking about is the penchant for commercial programmers with any sense of integrity to document their code with clear, concise, and accurate comments. (Although we never trust comments in lieu of the actual code.) Also he is talking about writing the logic in a straightforward and unambiguous style so that subsequent programmers can maintain it after you have left. And he is section he is pointing out is what is really criminal: intentional or unintentional, the algorithm that the guy wrote omitted data that did not fit within certain bounds.

Hardly encouraging to see something like that in code that is supposed to generate simulations of real world events. It's kind of like throwing out all of the checks you wrote for $200 or more when you reconcile your bank account. Or only record expenses of $1,000 or over in your general ledger.

Pretty soon, you will bounce a check or your financial statements will not reflect the actual state of your business.

So I want to pore through this code myself and see these problems myself. I would like to be able to offer my professional opinion as time permits, but, until I  can find the time, there are a number of sites where folks have done this already and - if you are interested - you should check out this discussion on Free Republic or this one at the same site.

And there's a good - albeit long - dissection of the code below. It's the second in this gentleman's series but the most expository. Warning! Do not watch this while operating heavy machinery. I recommend this for serious programmers and/or people with no life like myself. But the net of all this is that it's pretty damning.