Slashdot Mirror
NASA Prepares to Open Source Code
comforteagle writes "According to this story at O'Reilly, NASA is looking for approval for their own open source license. The NASA submitter (lawyer of course) states that none of the current licenses meet their needs, but more interesting is that NASA needs a license at all. It makes one wonder what we, and other space agencies, might see coming out off NASA. It's also nice to see code that taxpayers paid for anyway being released for their use too. There must be at least one slashdotter who could dream up a use for NASA software. X Prize participants maybe?"
Not likely to be very workable. NASA can't release a lot of stuff because of ITAR restrictions. The US of A treats most space related items as being ITAR Restricted.
For those asleep at the keyboard, ITAR is International Traffic in Arms Regulations.
For example, check out Flight Linux:
http://flightlinux.gsfc.nasa.gov/
You'll note that even though required by the GPL, NASA refuses to release the sources because of ITAR prohibitions.
Move along, there is nothing to see here.
This was originally developed while he (Gary Riley) worked for NASA at the Johnson Space Cener. It was available in source form since before I started working with it in 1993.
For instance, nedit, a great editor for people coming from Windows/Mac, was developed by Fermilab, a particle physics laboratory.
U.S. Code Title 17 Chapter 1 Section 101: and from Section 105: Although I will say that NASA seems to act like it owns the copyright on the images it produces.
Seastead this.
For over 30 years NASA code was available through a program called COSMIC which was administered at the University of Georgia.
http://www.cosmic.uga.edu/
In fact for awhile they operated out of one of the many buildings previously occupied by the 40 Watt Club
Since 1998 the code has been available through the Open Channel Foundation
http://www.openchannelfoundation.org/cosmic/
If you're refering to Maestro, that is not the code I was refering to. If you look on this page, you will see this picture, which shows some pretty darn good 3d processing for only using a few still cameras! And if you look here, you will find this picture which shows a representation of what the Spirit rover's software uses to find its own way, without need of constant instruction from Earth. Pretty good software, if you ask me! The public Maestro program is pretty slow, but so is Java in general for high-performance applications. Maestro doesn't actually generate the 3d range information from the raw images, it just displays it (and apparently is used to figure out the rover's schedule of stuff to do).
Out of curiosity, which NASA image processing software are you referring to?
Specifically, I am interested in code that can perform automated image mosaicing, also automated registration of images obtained through different modalities and code that will allow unsupervised k-means and/or ISODATA image classification/clustering of multispectral images.
Visit Jonesblog and say hello.
Please contact a NASA center and start asking around about doing joint research. NASA has a lot of research funding that requires joint work with a university, but you've got to hook up with the right researcher to get it. Every NASA center has an office that will help you find the people most likely to help you.
Believe me, if you need that image processing code and you are a university, a joint research agreement will get you a lot of help. If you need some other kind of code, and you think NASA has it, start calling around! It may be a bit of work, but you'll be surprised how eager many NASA researchers are to work with you.
But code quality aside, what about applications elsewhere? NASA's codebase presumably does a wide variety of things in addition to running gazillion-ly redundant life support on the space shuttle. Think about all the design and testing it does of hardware, the software it writes for image processing and signal analysis, running the deep space network. How about making models of satellite structural integrity? Surely something useful -- although it might take someone within the field to realize the similarity between a problem they face and one NASA has already solved.
And, of course, scientists love to write their own tools for text editing, data analysis (often these are incredibly powerful and extendable -- naturally more so than, say, commercial software products which remain close-sourced), collaboration software, yadda yadda ad infinitum
Protect your liberties. Donate to the ACLU
This is probably the package you are looking for: Integrated Software for Imagers and Spectrometers. It's been distributed freely for many years, and is, indeed, public domain. It's funded by NASA for use by NASA-funded researchers and the planetary science community in general.
Note, Isis 3.0 has not been released yet, look for the beta in coming months - look at Isis 2.1 for the stable release. Download/Install instructions are on this page: Isis 2.1 Installation Guide.
Automatic mosaicking is generally done using the spacecraft positioning information. Automatic registration? It doesn't exist (yet). Registration involves varying levels of human intervention, and when some level of automation is achieved, it's mission-specific and under special circumstances. Isis is primarily a cartographic package - IDL is generally used for statistical work.
Another image processing package that's public domain is USGS MIPS. It's a (non-NASA) terrestrial image processing package that evolved from the same roots as Isis, so you'll find it has many of the same capabilities.
I don't know what other NASA packages there might be out there like this, if there are any. I'll ask around.
A lot of it is "free", if only you can get it. The problem is that it costs agencies (and hence taxpayers) time and money to distribute, so there is no incentive to do so. For instance, government publications cost money, to defray the cost of producing and distributing them to the public. IMHO it makes sense for this cost to be paid by those who use the material, rather than by a lot of taxpayers who have no interest in it.
As for the rest, well, that's why we have the Freedom of Information Act. If you want source code for the accounting system for the Bureau of Public Works, put in an FOIA request and they'll either give it to you (for the cost of distribution, I guess) or give you a good reason why they won't.
NPR is not a government agency, but a private nonprofit organization, so your questions don't apply to it. However, even if it were, the government tends to use "industry standard" formats, and Real could certainly be considered that. As an example, all the forms on the IRS web site are in PDF, and they recommend (free but commercial) Acrobat Reader for viewing. Probably a lot of other files are available as Word documents, since that's how they are produced.
Moderators:
This is NOT a troll... the Skunkworks was the REALLY COOL department of Lockheed Martin that created all of their super-secret, very cool stuff.
Check out this link for some of the books written about them.
Personally, I'd be VERY interested in some of their code.
$0.02 (CDN)
It seems to be pretty clear to me:
I'm having a heck of a time finding what NASA's position is on giving the code away. Issues presented have been code security (what if my login page can be hacked?) and contractor vs. government ownership (but the contractor got paid for the work, right?).
Still working through the process, hoping the code will see the light of day before it becomes obsolete/irrelevant.