Enigma code broken

[The Duke]

Well it may of been cracked a wee while back, but AI just done it from scratch, including learning German.  it seems the only parameter it was given was its German.  then it basically tried evey given letter using statistical analysis to figure the German phrase.

It took a whole sub 13 mins, using 2k cloud servers.

I bet Turing would of like to play with this.

http://www.appy-geek.com/Web/ArticleWeb.aspx?regionid=4&articleid=125055265&source=clipper%2B

[blacklion1725]

Pah – that looks like the nancy-boy three-wheel Army Enigma – not a patch on the 4 wheel Navy version. Child’s play (I’m joking of course).

Off to Bletchley Park in the not to distant – so good topic Duke.

[RSB]

OK but I am guessing the guys/girls who created what ever app/program had been there at the time would of been peeing in there knickers. There is a lot behind the story/facts/achievements and it’s in my very own opinion twisted that anyone would want to come along and say “Look, we did it in 13 mins” whether it is just to show what is possible or not.

Not a go you SGB just a rant at  app makers and programmers trying to prove there own worth either by releasing pointless apps or diminishing other peoples work.

Seriously, some one had to pay for that super computer/cloud computer!

Americans: Over Sexed, Over Payed and Over here, Wat Wat!

[The Duke]

BL, it was the navel 4 cog. 56billion type 2.

Lee, I’m not sure why your down on this, it was shown at the imperial War museum, and was created but some real data security  scientists that are connected to enigma historical project.

It is meant to 1. Highlight how far we have come, since those small first steps. (small in computing terms, huge in brain/brain although terms) and 2.to highlight the power of AI.

Today is very similar in the computing journey then was the start of computing (disregarding the Babbage engine of course), today we are on the cusp of AI taking over.

I think this was a great demo.

Also rewind a few year before WW2, and the press then was saying the same rubbish about the you h of that day, just like they was pre ww1, and can only imagine, evey generation before.

Entre any, of our militery institutions, and they are full of great brave people.

Look up Anders Larson, a hero of mine, (both SAS and SBS fight over him), some of his biographies talk of how normal people can do anything when they have to. I also served in the RM  group named after the place of his death (I can’t spell it) camatchio ( lake/town in Italy). He also talks alot about the amount of drugs (bennies), commandos was issued and ate back then.

They was the in dance of hall drug of the 30s, a speed pill. Yes even youth back then took recreation drugs. Point is nothing has change.

Iirc cocaine was not long band from chemist sales around that time. So same issue then as now, opium, coke and speed. We now have lsd and hanks to the cia.

[Drezha]

The article says it was a marketing stunt by an AI company and Digital Ocean, so that explains who paid!

bl – it says it was the four rotor version that was broken within the writeup 😉

Interesting to see if done so quickly! I remember taking part in a distributed computing project that was using brute force to crack codes – it looks like it’s still going!

"Everything looks interesting until you do it. Then you find it’s just another job" - Terry Pratchett

[PlaneMan]

Drez, that project looks ancient in computing terms!?

[The Duke]

PlaneMan wrote:

Drez, that project looks ancient in computing terms!?

Turing broke his teeth on that project ?

[blacklion1725]

Well that’ll teach me to read it all before I post 😉

From memory I think one big breakthrough was that ze germans weren’t overly thorough in with the encryption of the daily (Enigma) code change notification, so while they changed the almost-uncrackable wheel settings every day, the protocol they used to send these settings was nowhere near as robust.

In any case no surprise that today’s tech can rinse yesterdays.

This side of the war is just as interesting as the military though, painful decisions around letting some convoys get sunk (and possibly the bombing of Coventry)? to hide just how deep we were in to the Nazi ciphers.

[The Duke]

From memory also, Im sure I heard or read accounts from the code breakers, that the signallers would open up each message the same, again out of lazyness.

Iirc days date, and to whichever general was in the area, each would start with (or something like that) . So the crackers would plug in These known knows first,  which could take days of the cracking times.

If you have a known set of say 10 letters, it’s means you could usually make very good guesses at filling in the blanks.

So you could either brake the cypher or break the message. Whichever came first.

As I said from memory, a long one at that. But I do love my WW2 history. Fascinating time.

[Bob Williams]

The two words that ended every Enigma message were the same, which was what eventually led to cracking the 3-rotor. Bletchley had an early version of that from the Polish Intelligence Service, and received it just in time in September 1939. The 3-rotor crack led to the 4-rotor crack, after a huge amount of work by the Bletchley team. The words?-

“Heil Hitler!”

Without Turing, the War would have lasted perhaps another two years, maybe longer. Without the whole Bletchley team, we would never have beaten the U-Boats in time and the UK would have been starved out of the War. We would have been forced to surrender, no matter what Churchill said. So much of the war’s crucial battles, in every theatre, hinged upon decoding that insanely clever machine. Even the Kursk tank battle, when the Red Army won the victory which started it rolling to Berlin. The Russian generals were given intelligence about the German Order of Battle, which came from Enigma intercepts. They were not ever told about the source, just that it was British Intelligence. Fortunately they acted upon it: something they rarely did with Intelligence reports from sources outside their own.

We owe that vanishingly-small band of old people so much, and the shade of Alan Turing more, considering what was done to him.

BL: enjoy Betchley, I have been with No.1 grandson. I advise looking for a B&B or cheap hotel, take two days. To see all of both the Museum and  TNMOC in one day, is impossible. But it is absolutely brilliant. Any younger people you can take with you, will make your own experience even better. If there is one of the very old guys or gals there, ask to shake a hand and tell them how proud we all are of them. There are very few left.

When the Thought Police arrive at your door, think -
I'm out.

[Drezha]

BTW, since The Imitation Game, the place has had a large injection of cash and is now even more worth visiting. The new section on the WRENS was interesting, as my Gran was one (though never at Bletchley as far as I know). I visited it prior to the film whilst at Uni doing my PhD and then more recently within the last 6 months or so and the money that’s gone in to it has been a breath of fresh air for them.

"Everything looks interesting until you do it. Then you find it’s just another job" - Terry Pratchett

[The Duke]

That’s great news Chris. I recall us talking about the place about 4 years ago (could of been 10 the way keep time) on mm, at the time they was raising money as it was on the verge of closer iirc.

One of the things I love about France, is its that big with a small population, they never knock anything down. It’s a wealth of old castles (better than any I’ve seen in Wales), nepolionic forts, ww1, 2 and clod war bunkers,  All there,  all funded and cheap to visit.

Iirc  11+ yo wad €5 under 11 free. Was the defacto price, non of this £20 an adult/15 for kids over 5 nonsense we have here, that prices most our of not only a great few hours, but a great few that could change the direction of a kids life.

I do live France, especially the northen half, it’s like Wales, with better weather and cheaper attractions. Just stay clear of Paris. Well like London City, it’s good for two nights max. Then you need to get out and explore.

[Bob Williams]

It seems that none of the 4 movies about the Enigma story, have told the real truth, although as you might expect, Hollywood tells the biggest porkies, including the outright calumny that a US submarine discovered the machine that helped solve its code:

http://tinyurl.com/ycdsnaa7

Even the most closely accurate movie, “The Imitation Game”, was wrong in parts. Including the fact that Alan Turing did not build Colossus, which was in fact not even built at Bletchley.

I mentioned Poish input before, but I forgot to tell the whole truth: that a Polish cypher team broke the first enigma code, before anyone else, including Bletchley.

http://tinyurl.com/nudklta

History, they say, is written by the winners. No, it’s written by those whose research is often carried out imperfectly.

 

When the Thought Police arrive at your door, think -
I'm out.

[JayCeeDee]

On a slight aside, but relevant – I was watching Guy Martin vs the Robot Car on C4 last week and they were telling you about early AI and one of Turing’s colleague’s supposedly started AI with an electronic tortoise that had a photocell and followed a torch shone at/in front of it.

 

The next real advance in automotive AI was at the TRRL – they had a 1971 MKII Cortina that drove around the circuit without a driver just following a white line and stopping when it reached a box. Apparently funding was withdrawn because it cost too much!! What would a 30 year head start on autonomous cars have been worth to the UK??

 

The programme is well worth a watch on CatchUp.

[The Duke]

Probably the same as those jet engine designs, atomic designs and all the other good stuff we bribed the US with to join us in the figjt.

Well without their help they may of helped Germany more.

The computing one to is a mystery. We was way ahead, then just let it stagnate for decades. That was probably in the contract for the US involvement.

[Bob Williams]

Richard and EdP can probably inform you more about early British computing and the failure to follow it up, Steve.

Tell us about the way it was ignored, fellas.

When the Thought Police arrive at your door, think -
I'm out.

[Ed P]

Blame Churchill and SIS for our computing failures. SIS convinced Churchill that it was vital to do three things:

a) Allow Enigma machines to ‘escape’ to other countries especially the USSR. (Third world countries were still using Enigma in the 80s)

b) Totally conceal British successes in breaking Enigma et al.

c) Most importantly Sigint was to be in the ultra-top secret class, and this has really only started to emerge from the shadows in the last ten years or so.

As a result not only was everyone sworn to life-time secrecy, but every bit of equipment associated with Bletchely was broken into little bits, and all Civil Servants etc discouraged from supporting fledgling computer projects. It was probably this deliberate disinterest that starved Ferranti of the funds it needed to push forward the projects that it had with the Government.

It did not totally succeed as even the Lyons Corner House tea-shop chain had the first(?) commercial IT set-up. In fact I learned to program on a LEO.

But unfortunately Churchill’s attitude pervaded the Civil Service. Were Bruce still around on the forum he could probably tell some stories about the way Top Civil Servants took pride in their ignorance of computers and technology.

They all share the blame for us moving from the heady days of Victorian technology leadership to our also ran status today.

[Richard]

Leo was an interesting project. I was taught maths for a while by a chap who had driven himself mad working on programming the thing. Well programming it was not the whole story, he started work trying to test the programmes as part of a team, the only issue was that the machine was not available because it was late arriving, of course. They worked the programme with pencil and paper. He left and became a disillusioned teacher. He had to move on again when his child was born severely disabled and they needed to find suitable support.

Another project was the Elliot automation computer controlled machine tools, I believe it was called the Elliot Auto-code 803. Quite what happened to that one is not in my memory banks but Elliot disappeared like all things.

I cannot confirm the Churchill story except to say that the original Colossus was broken up and the code breaking work was suppressed, I can understand why. However in the 1950s as those stories show there was some effort put into the nascent civilian computer effort. However Churchill was out in 1946 and the country was then under different management. It would be easy, dead easy to criticise the choices of the day. 1947 was a terrible winter and the imperatives were to give returning troops work and to try to get the country producing again but with no money with which to do so. Churchill had started the studies for the new national health in I think 1943which blossomed forth in 1948 under the new government. It needed staff to run it and one of the early objectives was to help ensure full employment. Overall the drive then was to get production and employment started, with export or die being the main driver. Through the late 1940s and early 1950s that was all that mattered. Sadly it was all done on clapped out old hardware, while I visited new factories making new stuff in the 1960s e.g. Mullard churning out transistors to replace valves it was still a cottage industry with doubtful yields –

Question: How do you decide what to make?

Answer: We make them and then test them, then we see what we need then we sort them into different types. those that fail testing go into transistor clocks.

Go down the road, or up to the Midlands and the story was worse, even by the 1960s engineering was still using machines made in the 1800s, OK the late 1800s e.g. 1890 etc. to supply parts to the car and lorry industry. Ship building was based on hopelessly out of date sites forced into hopelessly uneconomic ways of working. We probably had the worlds best riveters and boiler makers. Craftsmen making riveted ships that no one wanted for a price no one would pay. Everyone else had new purpose built new yards making welded ships. We would rather build the worlds most complicated steam engines that required large amounts of labour to keep them running while the world was moving or had moved to other motive power sources. New stuff took far less labour to keep it working often a quarter the manning levels. (This did not stop the likes of Argentina ruining their railways mind you.)

Some ideas failed for the well known reasons of bloody mindedness, with investment cash always short and the risk of long disputes over new working methods who wanted to throw money at long shot items that were sold as replacing boring work? Any one remember Wilson’s white hot technological revolution. I do, it delayed a whole load of essential developments in the communications world, a long boring tear inducing story of failure to understand. GEC, before they threw all the money into a pit marked do not go there. They set up a research and computer centre that they could only use for about 10% of its capacity. They decided to rent out spare capacity. I think they only had one or two customers before those customers filled all the spare capacity. I asked why only a few of the links were being used, they sheepishly said that their customer(s) used the machine for far more challenging stuff than they did and filled it up.

Where did we go wrong? My answer would be everywhere. There was frankly no single point of failure. A lack of imagination, a lack of flexibility, a lack of the right training, not understanding the way things needed to change, rigid mindsets, shortages of funds. Do not play mix and match just picking a few, they all applied plus a few more. Things had to break before anyone could stand a hope of rebuilding. I heard someone speak about alcoholics, they said do not persuade them to stop, buy them a drink. Then they can hit the ground while they are still in with a chance of recovery and, more to the point, can see their life was going wrong. Delay that vital step until they are too damaged to pull back and they are lost. So it was for our industry, subsidies kept no hope industries running while starving better ideas of needed funds. What little money there was got thrown at a string of flops, Linwood and Rootes, De Lorian et al.  There were other flops in Wales which tried and failed to create jobs. The civil service moved tasks to northern England, so thousands of female clerks type positions were supposed to mop up the unemployed from heavy industry, that was bound to work not. The damned jobs should have been data processing based years ago and built on a vibrant computing industry, not on HB pencils, but that would not create thousands of short term tasks.

Colour me bitter by all means.

[Ed P]

I thought that Inmos (an eventual success sold down the river by Thatcher et seq.) had disappeared but bits of it still survive as wafer fab shops – there is still one in Newport. Hopefully Brexit and loss of development grants will not kill it, as we need the homegrown SoCs that Inmos pioneered now more than ever. (transputer = SoC)

[Bob Williams]

EdP and Richard, I knew what I was doing when I called for your input. Thank you gentlemen, there can be nothing further to add.

When the Thought Police arrive at your door, think -
I'm out.

[Author]

Posts