I guess they're trying to seem responsive because the IPCC head didn't when this first came up. And because there's a media feeding frenzy over this (well, in the UK at least) where any bogus claim of a global warming science error can get an airing.

Next time they just have to remind all their Working Group II and III authors and editors that those guidelines they've got on how to use 'gray' literature are actually meant to be followed. And they could use some routines to make sure that, say, a glacier expert looks over everything that is said about glaciers in those 3,000 pages, and not just in the glacier chapter. But it's an open, volunteer-based process, so I guess there are limits to how tight a ship they can run.

The IPCC doesn't do research. It "reviews and assesses" the science in order to "provide the world with a clear scientific view on the current state of climate change and its potential environmental and socio-economic consequences".
http://www.ipcc.ch/organization/organization.htm

The 'deniers' didn't find the Himalaya glacier error. It was found by a glacier expert, Georg Kaser, who happens to be one of the lead authors of the snow-and-ice chapter in vol. 1 of the IPCC report, which deals with the physical science basis for man-made climate change. (No errors found there.) The error was in vol. 2, which deals with the impacts of climate change, way down on p. 493 in a 'case study' inside the chapter on Asia, which apparently was not reviewed by any glacier expert.

The other error -- regarding the percentage of the Netherlands that is below sea level -- came from no lesser a source than the Dutch government (oops).

Speaking of which, the update just failed to work on my PPC (G4 Powerbook, running Leopard, so far updated to Java 1.5.0_16). Anyone else have trouble? Any clues?

Author-side payment is one funding model (or one element of funding models) for open access journals, pioneered by PLoS and New Journal of Physics. Articles are made freely available on the internet. I have seen estimates placing the cost of publishing a scientific article well above $1900. Whatever the truth of that -- publishers have to recoup their costs somewhere and the traditional model of selling subscriptions (at comparably breathtaking rates) to research libraries is slow, restrictive, and also favors rich institutions in rich countries. Under many (full) open access models, at least author charges are waived for Third World researchers.

Works pretty well with the Firefox add-on Downloadhelper, in my experience.

Except that coelacanths aren't dinosaurs, sure. The scary bit? Wikipedia says "individual coelacanths may live as long as 80 to 100 years". Good luck with that move from IE6...

There must be some business logic to Amazon's confining their ebook sales to their own format, their own device, their own network, and their own home country. Don't know what it is though.

I'd be buying my ebooks from Amazon if I possibly could. But I can't. They don't distribute over the Internet so I can't download to my preferred device. They don't make Whispernet available outside the US so there's no incentive for me to ditch my preferred device for a Kindle.

No it doesn't. Greenhouse gases do keep in heat radiation. They don't stop visible light coming in from space, and they don't stop reflected visible light going back out. The visible light that is not reflected, heats the surface, which then radiates heat in the infrared, which is what is absorbed by greenhouse gases. Painting roofs white would mean more visible light is reflected back out of the atmosphere, not more heat. Do try getting a very basic grasp of the physics involved before ranting.

Check out Prince (princexml.com), a CSS-based typesetting engine that outputs PDF from HTML or any valid XML. It's commercial, but there's a free version for personal use. It currently doesn't do math at all well, so it's no contender whatsoever in the present discussion, but it's definitely something to watch for separating content from presentation with ease of use, commonly used formats, and a small footprint.