Second World War Code-cracking Computing Hero Colossus Turns 70

DW100 writes "The Colossus computer that helped the Allies crack messages sent by the Nazis during the Second World War has celebrated its 70th birthday. The machine was a pioneering feat of engineering, able to read 5,000 characters a second to help the team at Bletchley Park crack the German's Lorenz code in rapid time. This helped the Allies gather vital information on the Nazi's plans, and is credited with helping end the war effort early, saving millions of lives."

Comments

In other news...

[kry73n]

the slashdot beta sucks

Bomba kryptologiczna

[Savage-Rabbit]

Colossus, Alan Turing and the geniuses who helped design it, have been key to the development of subsequent fantastic advances in computer technology and marvels that have forever changed the face of the world, such as AOL CDs, Angry Birds and Facebook.

Alan Turing was a indeed a colossus but he didn't crack the enigma code. He didn't even lay a lot of of the ground work for designing this machine, it was a team of mathematicians working for Polish military intelligence after Polish and French spooks had gained access various data concerning Enigma that included inspecting a working copy of an enigma machine. Their names were Marian Rejewski, Jerzy Róycki and Henryk Zygalski and they reverse engineered the Enigma based on this material using mathematics and created what they called the 'bomba kryptologiczna'. The famous Colossus was a 'substantial develpment' from this device. What Alan Turing and Co. did was crack the improved enigma machines (still a daunting task) who had been upgraded in 1938-39, but he and and his team stood on the shoulders of those three polish mathematicians. The British are very keen to take sole credit for cracking Enigma but they got a whole helluva lot of help from Poland and France and as a German I'd like it to be crystal clear to the world who exactly it was that kicked our cryptographic ass :-)

Re:Bomba kryptologiczna

[Vanders]

The Colossus wasn't used to crack Enigma: Bletchley has simple electro-mechanical machines (Bombes) for that. Colossus was used to crack Lorenze, which was an entirely different cipher. The basis for the software that ran on the Colossus was basically Alan Turings work on cryptanalysis, and of course it was also Turing complete. The actual design however was almost entirely the work of Tommy Flowers; a post office telephone engineer.

Re:Bomba kryptologiczna

[Savage-Rabbit]

And there is the fact that US didn't capture a German Navy Enigma from a submarine as portrayed by Hollywood.

Sure the Brits owe a lot to the Polish who nicked an enigma but as has been said the advanced machine was a really different beast to crack. Also the German Navy changed their codes at a critical time of the war. Without the people at BP a lot more lives would have been lost, a great number of them US troops on their way to Europe.
My Mother worked there for two years (1943-45). She never said anything about her work until the late 1970's so naturally I won't hear anything said against the people who did that almost impossible job.

It's been a while since I read about this so I may be shaky on details but I don't think the Poles nicked an Enigma but they did buy one. There were several variants of Enigma divided into two different classes, the commercial and the military units. You could buy a commercial grade unit quite without restrictions so the Poles did that just to get an idea of how it worked. Later on they bribed a Nazi customs official to get access to a military Enigma and inspect it because stealing it would have alerted the Nazis that the Enigma system had been compromised. The French also contributed data and that led to the construction of the 'Bomba'. When the Germans added more rotors to the Enigma in 1938 it massively increased the magnitude of the required decryption effort and the Poles didn't have the resources to construct the requisite machinery. That's where the British became involved.

Re:Dice just killed Slashdot

[martin-boundary]

Actually I blame the developers. Clearly they are modern "app" developers, as opposed to us old school application developers, from an era when "application" didn't mean a glorified web page drenched in javascript toys.

Any application developer worth anything knows about Model-View-Controller and can separate the underlying data model interaction protocols from the view presented to the user.

There is no excuse whatsoever for the loss of any existing features that are found in classic, nor is there any technical reason whatsoever why there has to be a migration to a single "new" site to keep up with the times. The slashdot website is just a view into the comment and stories database, and there should be many views for everybody to choose their preferred one at any time, including the "beta" one as just one of them. In fact, if slashdot published a reasonable API there would be plenty of low digit users who could whip up a sane interface before breakfast.

There would have been no victory without the resou

[qubex]

That’s a totally dubious opinion misstated as fact.

Without American materiel (lend/lease ships, tanks, bomber aircraft) and manpower (D-Day landings, continental fighting, naval convoys) the war effort would have been almost inevitably lost. This does not mean that the UK mightn’t have eked out a long-term stalemate and perhaps even an uneasy truce, but the defeat of Nazi Germany would have been out of the question. What ultimately defeated Germany was not the war on two fronts, but an expensive, resource-intensive war on two fronts that exceeded the country’s ability to regenerate. Without the virtually bottomless reserves of resources provided by the USA, the USSR would have been eventually brought to heel, and the West would have followed suit.

The USA was pivotal.

Correct.

[qubex]

The Colossus was useless at decrypting Enigma traffic: that was handled by the electronic bombes.

Colossus was constructed to break Lorenz/Tunny traffic: a much more advanced system designed for encrypting teleprinter five-bit Baudot-code teleprinter transmissions. Dilettantes will harp on Tunny’s greater number of rotors, but it was a far more radical departure than might at first appear. As many subsequent stream-ciphers, Tunny XORed cleartext to a cryptostream. Amongst other things, that meant that there was no restriction against a character in the ciphertext being the same as the corresponding character in the cleartext, a flaw which allowed skilled cryptographers to infer what might, conceivably, be contained within a given stretch of text.

Two sets of ‘wheels’ were summed independently to a five-bit cleartext word. One set was advanced on every word and one advanced only if another wheel’s value was !FALSE (this wheel itself advanced on every word). This meant, amongst other things, that sometimes part of the keystream did not increment, and this in turn had a discernible effect upon the statistical distribution of the difference between successive ciphertext words.

Reconstructing the keystream from these distributions is how Tunny was broken, and that is the task that Colossus was designed to automate. (Mumbling about Colossus’ Turing-Completeness is fundamentally ill-posed, as no machine has the infinite memory capacity envisioned by Turing. I will however emphasise that Colossus lacked a stored program facility, a concept that was only developed much later.)