Slashdot Mirror

I've posted this before, but heres a non-comprehensive list of the raw data: knock yourself out.

Data:
NOAA NCDC: http://www.ncdc.noaa.gov/oa/ncdc.html [noaa.gov]
NOAA sattelite data: http://www.class.noaa.gov/ [noaa.gov]
ARM data: http://www.archive.arm.gov/armlogin/login.jsp [arm.gov]
NASA GISS data: http://data.giss.nasa.gov/ [nasa.gov]
NCAR data: http://www.ncar.ucar.edu/tools/datasets/ [ucar.edu]

Models:
NASA GISS GCMs: http://www.giss.nasa.gov/tools/ [nasa.gov]
NCAR models: http://www.ncar.ucar.edu/tools/models/ [ucar.edu] ...

That being said - I think this whole 'give me the raw data' thing is just a big red herring. What are you going to do with it when you get it? Do you understand all the sources of the data, their biases and errors? Do you understand how they calculated some of the parametrizations in the models for bulk cloud parameters? Do you think its just a matter of plugging the data into a code and getting a Yes/No result about global warming? Come on! Theres a reason this stuff generally requires a PhD or years of experience to understand. But there it is, lots of data. Now shut up about the data or methods being unavailable.

The data is all there (especially from US sources). Here's a very non-comprehensive list:

Data:
NOAA NCDC: http://www.ncdc.noaa.gov/oa/ncdc.html
NOAA sattelite data: http://www.class.noaa.gov/
ARM data: http://www.archive.arm.gov/armlogin/login.jsp
NASA GISS data: http://data.giss.nasa.gov/
NCAR data: http://www.ncar.ucar.edu/tools/datasets/

Models:
NASA GISS GCMs: http://www.giss.nasa.gov/tools/
NCAR models: http://www.ncar.ucar.edu/tools/models/ ...

Extremely non-comprehensive list based on sources I've used:

Raw Data:
All of NOAA's satellite data http://www.class.noaa.gov/
NCAR's Data http://cdp.ucar.edu/home/home.htm
ARM data http://www.archive.arm.gov/armlogin/login.jsp
NASA data http://data.giss.nasa.gov/

Models:
NASA GISS GCMs http://www.giss.nasa.gov/tools/
NCAR models http://www.ncar.ucar.edu/tools/models/

Storms like those encountered by the flight have very strong up and down drafts.
The warning issued by air bus about flight speed indicates the plane was flying close to stall.

A very strong down draft + slow flight == crash.
As the recovery for a crash is a dive. But in a strong downdraft your fall speed can't get you the lift needed to overcome the stall.

Downdrafts are serious business
http://www.arm.gov/publications/proceedings/conf09/extended_abs/xu_km.pdf

Hmm. Thats the same frequency as the radar used to track clouds

W-band ARM Cloud Radar (WACR)

So it reflects off of water droplets.

Hmm. I wonder if fog machines will be de reguire for mass protests in the near future.

Personally I'd think that would look a lot more hip than space blanket ponchos

Journal inaccessibility is one roadblock for researchers, another is easy and quick access to research data. What if you were able to see research data as it's being produced? That combined with easy access to journal articles, like supplied by the physics eprint archive, would certainly improve the research environment. One of the NSDL collections, the Atmospheric Visualization Collection, allow users to visually explore near real time atmospheric research data for just this purpose.

Okay, ignoring the ad-hominem "blatantly false and jingoistic" . . .

Sorry about that, its just sometimes people seem to just be as guilty of spreading FUD as the so-called "evil" corporations that it gets exasperating.

I apologise for those comments.

Now . . . you're saying, if I understand, that the NSA's SE Linux is just hacking the Linux kernel to put in some stuff that's been talked about and even done in other OSes for years? And stuff that isn't even all that novel for Linux?

Yes and Yes. Actually what regular Linux is implementing (which is different from what the NSA is doing with SE Linux) is POSIX 1.e capabilities or "priviledges" which involves splitting up the permissions typically given to the root user (e.g. can connect to ports under 1024, can mount kernel modules, can change ownership of files, etc) into discrete entities that can be apportioned to other users and processes. This was something that the POSIX folks tried to agree on in the eighties (or is it seventies) but never came to an agreement on how best to implement it. Check out the Linux Capabilities FAQ for more information.

The NSA is working on "true capabilities" which is being able to grant and revoke extremely granular permissions to all objects/entities in the system. This concept is similar to java.policy files being maintained for every entity in the system. Making sure that policies can be tracked in such a manner that they are revokable is the most difficult part (e.g. if I lose permissions to connect on a certain port or write to a certain file, then every process or file that I've created should lose those permissions as well).

--