Searchable PDF created for NASA GISS FOIA documents

As we reported earlier:

Newly released FOIA’d emails from Hansen and GISS staffers show disagreement over 1998-1934 U.S. temperature ranking
click for a larger image

Now thanks to the efforts of Richard Henry Lee, a searchable PDF document of those files has been set up.

Download it here:

NASA_FOIA_docs_OCR (PDF  – warning, 10 MB file, may take awhile on slow connections)

Once opened, use the “binoculars” icon in Adobe reader to search on keywords. To minimize data traffic, both on your end and mine, I suggest you save a local copy and do the search there. Otherwise you’ll have to download it each time you follow the link, and that’s quite a time and data sink.

Other databases are also being setup, watch this space.

0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
January 18, 2010 1:55 pm

Thanks, it is always useful to have a searchable database.

a jones
January 18, 2010 1:55 pm

So once again it is the dedicated amateurs with their informal networks who outrun either the MSM or the authorities in discovering what there is to be found. I make no judgement on that at this stage.
Time will tell.
Twenty years ago they were hailing the word processor as the new Gutenberg that would change publishing.
They were only half right. It is T’intenet that is the truly wonderful thing: and it is changing the world.
For it can give tongue to the masses in a way not seen since printing and the gradual rise of literacy that followed.
My friends we live in interesting times.
And I for one welcome it.
Kindest Regards

January 18, 2010 2:04 pm

First we had the Ozone Hole, then overpopulation, then maybe something interesting – now we have fear of carbon and Islam. Both silly. We are made of carbon, so banning it we ban our own lives; and Islam is learning from the Japanese Kamikaze, desperate people reach for desperate measures. I think there was also the asteroid that was supposed to hit the Earth. And sun storms were to kill life on the planet, too.

James Chamberlain
January 18, 2010 2:13 pm

Date: 16 Aug 2007 18:16:22 -0400 From Gavin
“It may still be worth putting up a clean version of the adjustment program on the website in order to have something to point to in such cases.”
This statement scares me. Not being in programming, does “clean version” hold some proper definition? I read it as the “sanitized version.”

January 18, 2010 2:39 pm

now we have fear of carbon. No one’s afraid of carbon by itself. Just too much of it. Same is true of anything we eat or drink as well.

brad tittle
January 18, 2010 2:46 pm

“Clean” version means that you get rid of the many lines of commented out code that hasn’t done anything for years, but someone commented out and the next person didn’t know why so it was left only to have more code commented out, then new code added.
Quite often the code gets bloated this way. The best I ever did was turn 100 pages of code into 1. It didn’t require much brilliance to make it happen. All I had to do was say “Hmm, that code seems to have been cut and past 20 times. What if I just consolidated it…” When I started on that project, the distribution required 7 disks (floppies). When I finished, it required three, two of which were the libraries (FoxPro), to make the application work. Hindsight tells me this was probably a bad thing to have done. It sounds much better to say that you created an application that took 7 disks, than to say you made one that took 1.
I don’t want to defend Hansen et al too much, but “clean” probably isn’t a bad thing. Even the best code has hacks in it. Just the other day I had to add 7200 seconds to a timestamp. The first time the element calculated correctly, all subsequent calls were off by two hours. I did my best to comment it appropriately, but it looks like hell and makes absolutely no sense, but it works now, which is quite often what we programmers are trying to get happen most of the time.
That little hack may come back to bite me.

January 18, 2010 3:11 pm

This statement scares me. Not being in programming, does “clean version” hold some proper definition? I read it as the “sanitized version.”
It’s context dependent, but generally what programmers mean by “clean code” is code that has has had the cruft removed, i.e., dead code removed, formatted for legibility, seat of the pants make-it-work code replaced with better thought out solutions.
The real test, of course, is to take the published source and data and see if the output can be reproduced, without having to make any changes except for porting.

January 18, 2010 3:19 pm

My college C (the language, not the grade) professor told all of us that a lazy programmer was a good programmer. It’s better to do something in one line than 20.

January 18, 2010 4:18 pm

SteveK (14:39:36) :
Someone pointed out a substance that was one the most addictive substances in exsistance: Once you’ve used it you’re hooked for life. The withdrawal symptoms are terrible and once you have used it stopping will kill you. The substance? Oxygen.

Jenny A
January 18, 2010 5:41 pm

The most damning tidbit you can come up with is some quibbling over a tiny fraction of a degree difference between 1934 and 1998?
It’s really old news. The more you run with it, the more petty and pointless this whole endeavor appears.

Dave F
January 18, 2010 5:50 pm

Jenny A (17:41:49) :
BFD? Well, 1934 went down, 1998 went up. 1, 2, 3, how many times before the result was the desired one? Tell you what, I have this company that you should invest in. Just tell me what your ideal target rate of return is, and I’ll meet it. Just don’t ask me how, or try to make a withdrawal, ok? Minimum deposit of one trillion dollars. Do we have a BFDeal?

Roger Knights
January 18, 2010 7:20 pm

Jenny A (17:41:49) :
The most damning tidbit you can come up with is some quibbling over a tiny fraction of a degree difference between 1934 and 1998?
It’s really old news. The more you run with it, the more petty and pointless this whole endeavor appears.

I agree that mini-scandals like this, including Climategate, don’t affect “the science” very much, and that “our side” shouldn’t attach as much weight to them as it has been doing. What they do affect are “the scientists,” the persons whose judgment we laymen must take on trust. I.e., their importance lies in the intangible realm of suspiciousness and trustworthiness.
For instance, if you’ve detected the company treasurer slipping out $20 bills from the petty cash drawer from time to time, it arouses suspicion as to what bad habits he may be a slave to, and what he may be doing to the company books. It justifies calling in outside auditors to go over all his documents (including e-mails, etc.), review his practices, etc. Hyperventilating about the $20 loss involves a loss of perspective.
Our side should not be claiming that AGW has been debunked, but merely that it’s no longer as settled as it needs to be, given what’s at stake. This minimal claim is credible; anything more is controversial. Its implication is that panels of independent scientists (e.g., under the NAS) should oversee and examine all past and future work and reasoning in this domain, and take depositions from contrarians. The institutions that constitute organized clime, including the IPCC, lack sufficient objectivity and credibility for the task.

James Chamberlain
January 18, 2010 7:31 pm

Thank you Brad and Larry. That is exactly what I was wondering.

January 18, 2010 8:14 pm

What most strikes me from this series of mails is repeated claims (paraphrasing) as to ‘we don’t have any code we document what we do’. Then later on you see talk about posting ‘clean code’ on So is there or is there not code.
If there is code, are the people claiming there is none out of the loop, being disingenuous or flat out lying?
It makes the correspondents appear that the left hand and right hand not only don’t know what the other is doing, but don’t even know about the existence of their ‘opposite’.

January 19, 2010 1:55 am

Roger Knights (19:20:38) :

Smoking Frog
January 19, 2010 3:59 am

hoystory 15:19:49: Calling it “lazy” to do something in 1 line instead of 20 is at least questionable. (A) In many cases, the 20 lines accumulate having no obvious relationship to whether the programmer is lazy, and then it takes more work to reduce them to 1 line. Most often, the programmer who wrote the original 20 is the one who subsequently reduces them to 1. (B) In many cases, speaking now of 2 different programmers, something is done in 1 line instead of 20 because the programmer who did them in 1 is smarter than the programmer who did them in 20. This has nothing to do with laziness. In fact, the programmer who did them in 1 may have beaten his brains out figuring out how to do them in 1, and/or the programmer who did them in 20 may be lazy. (C) In some cases, something is done in 1 line instead of 20 by eliminating useful “functionality.” This might be lazy, or it might be that the programmer loves doing things in 1 line and is willing to sacrifice something useful to achieve it.

January 19, 2010 6:13 am

Can’t resist an observation:
Weather and climate are like “code”, there are zillions of “lines”. Mankind has been scratching his head for eons trying to figure it out. He started with simple things like “red sky at morning, sailor take warning etc.”, moved on to “thermometers”, and ‘barometers”, etc. Now we’re trying to “code” things to fit the facts. Me thinks we may be “cleaning” The Master’s Code too much? (Or just not seeing the value of all those “lines” we think are superfluous now –but which are critical to the “program”.) Like I said, just couldn’t resist:-)

Steve Richards
January 19, 2010 6:15 am

The problem with one line of code compared to 20 lines:
The one line of code may have been written by a genius but that one line of code may have to be supported for the next 20 years by people who do not have the time to understand the nuances of that one great line.
Give me a tidy 20 line than the clever one line any day.
You only look at the code when you try to understand it (ie now with climategate) or when you want to remove a subtle bug.
Do things step by step (20 lines) not in one big hit (1 line)

Roger Knights
January 19, 2010 11:26 am

Summing up, there are bad & good short code segments and bad & good long code segments — it all depends.

Zejith Themis
January 19, 2010 4:49 pm

Smoking Frog (03:59:01) :
Steve Richards (06:15:02) :
This is the (information) age-old dilemma of legibility vs. efficiency in code.
What’s good for the CPU is difficult for the human, and vice-versa.
The Paretian optimum is always difficult to find, and despite what the studies of “sustainable development” (is it just me or is that a quasi-oxymoron?) may think, not always a true optimum solution.
“Clean” is in this case a question of opinion. Is it human-legible or smooth processing? What programmer makes the comment and what’s their preference?
Full context is everything.

Zejith Themis
January 19, 2010 5:11 pm

ooo! ooo!
Re: my Zejith Themis (16:49:47) : to the Steve Richards (06:15:02) :
(catalogued) Source code comments should serve to avoid the problem of the “clever 1-line” being illegible.
Comment and keep all versions of source, especially when working on the public’s dime.
That should have been in my previous post, 2:00am posintg brain-bug there.

%d bloggers like this: