Posted anonymously, like saltwater taffy.

International Space Station Incorporates Columbia's Kermit Software Program

slap on a kermit and save the day
by Michael Larkin
New York - Dec 09, 2003
Created almost 25 years ago by Columbia's academic computing center to help manage the high demand on the University's mainframes, a software program known as Kermit has leapt all the way to the International Space Station where it is being used in a scientific experiment.
Designed to allow two different computer systems to interact, Kermit was used to solve a compatibility problem on the space station. Using two versions of program, one of which was modified specifically for NASA, an experimental device referred to as CLSM-2 can now share information with another computer on board the space station that transmits data back to earth.

"Kermit and Kermit 95 have been invaluable tools to improve our computing efficiency, both in development and in the final operational system," wrote Dave Hall, senior engineer, ZIN Technologies on Kermit's Columbia Web site.

The significance of Kermit is not entirely its invention or its inclusion in the state-of-the-art experiment, but its ability to evolve and to retain its viability in the always-expanding computer industry.

And as one of its creators admits, it was never imagined that Kermit would develop the way it did. "Nobody expected the protocol and software to become a worldwide de facto standard, but even if we had, there are not many things we would have done differently, except in choosing a name," said Frank da Cruz, a manager who has worked on the project since its inception. He recalled amusingly how a picture of the friendly green amphibian swayed his judgment when it came time to name the project.

According to da Cruz, Kermit was borne out of a project to alleviate the strain on the University's academic mainframe computers in the late 1970's, which could only provide 35KB of storage per student. Columbia employees developed a protocol to transfer information from the mainframes to floppy disks through microcomputers that were installed around the university. The first Kermit file transfer occurred in April 1981.

The introduction and the ensuing popularity of IBM's personal computer (PC) prompted the next stage in Kermit's evolution. The university adapted the Kermit protocol to address the PC's incompatibility with Columbia's other computers and released it in January 1983. The PC version proved widely popular and was the subject of books published in English, French, German and Japanese.

At the same time, Kermit programs were developed for minicomputers being used in several Columbia departments. Its popularity continued to grow through the mid-1980s, and by 1986, Kermit was well established at Columbia and a fixture at many other universities, government agencies and companies worldwide.

Through the years, hundreds of Kermit programs have been written at Columbia and elsewhere and distributed through the project. In the early 1990s Kermit software was engineered to handle Russian, Hebrew, Japanese, Polish and many other languages via both their traditional character sets and Unicode, the new Universal Character Set.

"At conferences in Europe, the Soviet Union, and Japan, we quickly came to appreciate the enormous demand for computer communication in diverse languages and writing systems, and worked to make it a reality," said da Cruz.

Kermit 95, which was created for Windows 95 and its successors, was licensed to universities such as Oxford, Harvard, Dartmouth, Princeton and the entire SUNY college system; and was bulk licensed to over 800 companies and government agencies worldwide.

Kermit was initially shared with other organizations at no cost, despite the fact that it used a great amount of resources to coordinate the writing of new programs to archive results and to distribute the software. But in 1986, the Kermit Project was formed and distribution fees were establish

You need to understand the differences.

zmodem is high performance single streaming large packet size negative-acknowledgement only protocol - it fails badly in noisy or lossy style of environments.

kermit is far more robust, can interoperate with various different systems of different character encoding, had adaptive retransmission, and can perform just as well as kermit under the right circumstances.

The BBS implementations of kermit were not as sophisticated as the protocol could be, and most BBS environments didn't need the kind of features that kermit had. kermit is also of the emacs style: it's not just a protocol by an entire interactive terminal in itself: scripts, command line, etc.

...the communications protocol that wants to be an operating system. Somebody (Frank) has too much time on his hands. Reminds me of EMACS. These programs are definitely not in the spirit of UNIX.

Boy, it takes me back to read the word "Kermit" when not related to a frog... I actually used to use that... but there was no Slashdot then to talk about it on.

And that's probably why they use it. One of Kermit's strengths was that it was so robust, you could use it over a link comprised of a squirrel talking to a chipmunk by way of two drunk beavers, and it'd still work.

Seriously, just about every mainframe I've used has supported Kermit, and it'd work when almost nothing else would. It was nice when a machine had Zmodem loaded, but they often didn't, but they always had Kermit, which is why I kept it around on my PC long after I ditched DOS in favor of Windows. I never did get Kermit95, though. Never understood why they went to a pay version.

Need reliability in an environment where your electrical equipment is being slowly dissolved by ionizing radiation?

You just made a great case for Kermit, but I feel it needs a bit of a summary:

in space, reliability is key

Often triply redundant systems are deployed, and their life expectancy is STILL 5-10 years at best.

it brought a smile to my face to imagine the old protocol from my BBS days (which was scorned in favor of Zmodem) being used on the greatest technological achievement of humankind."

You have got to be kidding me. Maybe maybe the moon race qualifies as the "greatest technological achievement of humankind" I have yet to hear of a single usefull discover onboard that (expensive) piece of low flying equipment.

Don't knock it, man... there are tabloids, and then there is the Weekly World News.

Q: What's the difference between the Weekly World News and the New York Times?

A: The Weekly World News tells you that it makes up its stories.