Turing, Tutte & Tunny - Computerphile

Professor Brailsford rounds up the whole Colossus affair, and explains how Turing actually played a small but significant part in Bill Tutte's work.
Colossus Playlist: • Bletchley Park (Coloss...
Enigma Playlist: • Alan Turing and Enigma
What Happens When I Press a Key?:
Choosing a Password:
/ computerphile
/ computer_phile
This video was filmed and edited by Sean Riley.
Computer Science at the University of Nottingham: bit.ly/nottscomputer
Computerphile is a sister project to Brady Haran's Numberphile. More at www.bradyharan.com

[@MrDaanjanssen]

Enough programming for today, time to watch a video on programming

[@asdfkjhlk34]

thats exactly why I'm here too

[@Howtheheckarehandleswit]

Don’t call me out like this

[@ninadsachania3652]

A 37 minute Computerphile is exactly what I need right now.

[@oneofspades]

Its Professor Brailsford so you know its a gold nugget.

[@foowashere]

Hear, hear

[@marrrtin]

Blown away with this. Super high clarity and fab animations. Really sets the record straight on who did what at Bletchley. Bravo Prof Brailsford.

[@artemonstrick]

I CAN LISTEN TO HIM FOR HOURS

[@parrotraiser6541]

The change in meaning of "computer" had a parallel earlier with "typewriter". It originally meant the person who used the machine, (what we now call a typist, or did until the machines fell out of use). One of G.K.Chesterton's characters "falls in love with a typewriter". I thought "I know this is a surrealist story, but that's getting really weird", but then I found out about the shift in meaning..

[@greeneightball]

Professor Brailsgord, you're an icon for the computing community and one hell of a story teller. Thank you for another excellent breakdown.

[@plapbandit]

Yes yes, 37 minutes of Brailsford get in

[@emanuelebonura783]

I imagine an old man sitting in Germany right now watching this video be like: "Scheisse so THAT's how they figured it out then..."

[@bennylofgren3208]

Emanuele Bonura Or worse, "oops... I probably shouldn't have sent that message twice without changing the settings... 😳😬"

[@williampodmore8862]

Emanuele Bonura

[@brainimp]

Reason Tommy Flowers never gets the credit like others is because he was "only" a Working Class East End Boy that gained his knowledge through an engineering Apprenticeship and doing a part time course in the Evenings, He wasn't from the elite universities with a privileged upbringing like a lot of his Peers and the journalists that later writ about them.

[@profdaveb6384]

Craig: I am sympathetic with the case that you put forward while also agreeing with the points made in the reply from Rusty S. Tommy's achievement of a degree via evening classes etc. showed just how talented and determined he was but having come via that route he didn't have the comforting fallback of being part of the Cambridge academic community (unlike Shaun Wylie, Alan Turing, Bill Tutte and Max Newman). In terms of formal honours Tommy was awarded an MBE; Turing got an OBE, Newman was offered an OBE but turned it down as being 'derisory' (!) . Here again the lower class of British Empire order, for Tommy, shows up yet again the Establishment's reaction to his working class origins. Incidentally, in terms of formal honours, the one who got nothing at all was Bill Tutte. However, his obituary records that he was immensely gratified (in the 1980s I think) to be made a Fellow of the Royal Society. .

[@1Passingthrew1]

Interesting and very true. The Establishment shows up, albeit in humour, in Yes Minister where Jim Hacker went to the London School of Economics unlike the Establishment Civil Servants. It also shows the struggle that Harrison had in Longitude the movie as a country carpenter getting his clocks recognised and being given the prize. Incidentally i think you meant peers meaning equals rather than piers.

[@b43xoit]

"Peers".

[@martijnheeroma5492]

Thank you Professor Brailsford, what a gift.

[@bryanl1984]

I wouldn't admit to that... ;p

[@davel8116]

If it's Brailsford, I'm watching.

[@donaldasayers]

I learnt recently that the Germans broke the British naval codes used in the early part of the war. Can we have some videos on the the Axis code breaking effort? There is very little online history of this.

[@risvegliato]

Exactly what I was just thinking! I have seen scores of videos of how the allies broke the enigma codes, etc, etc, but I have not seen one regarding the reverse. Someone must know the history...?

[@hennobrandsma4755]

The book ‘Hitler’s spies” has most of that history. In short, they didn’t achieve much against allied codes, a little against the Soviet ones.

[@MrHatoi]

@@hennobrandsma4755 Turns out soaking your messages in vodka doesn't hide them too well.

[@sikfaka1]

I would also like to learn about the progress, bottlenecks and breakthroughs the germans had in creating the enigma itself from an axis perepective

[@fortboy66]

Didn't the Germans capture quite a few British typex machines at Dunkirk? I read that somewhere?

[@trueriver1950]

When reading 1950's SF you often meet the term electronic computer. The last minute of this video explains why that term was used

[@jmyers9853]

i love this guy, his method of describing what went on is very informative

[@hirakmondal6174]

The moment I saw Prof.Brailsford, Colossus By Jack Copeland and the the Bletchley Park codebreakers, I knew i am in for a treat.. XD

[@_infinitedomain]

I could listen to this guy talk about anything, so engaging. Brilliant video.

[@CraigWest1]

Professor Brailsford is a true national treasure! I can listen to his presentations for hours.

[@TheSharkey22]

Marvelous. Thanks for all your time and effort. Keep 'em coming.

[@charlesmiller000]

Professor Brailsford is a gift that keeps on giving! Long & healthy life Prof.

[@PhilBoswell]

I'm so glad that these have been collected together in a playlist, as it's so long since the previous that I've forgotten half of it. At some point i shall have enough time to go back and binge the whole lot ^_^

[@gordonrichardson2972]

Very coherent summary, and useful elaboration on some details mentioned in previous videos in the series!

[@MindstabThrull]

Just noticed that CHI and PSI look like a visual tweak on X and Y. That's probably why they were chosen.

[@danielrazulay]

Seems to me they're both related to Euler.

[@lowwatthalo1654]

I'm out of my depth with parts of this but he manages to keep me engaged. The historical narrative keeps this fascinating & also shines a light on the unsung heroes.

[@josephwong2832]

What an epic story You can see this guys passion and enthusiasm of how important these techniques were for the times

[@edmungbean]

i am absolutely stunned. This was incredible.

[@brianmessemer2973]

I love that casual SOLVED Rubik's cube on the shelf. Legend, this man.

[@Snowboard4466]

I love this professor. I think half the views on this are just me watching it again and again.

[@garfieldfarkle]

This is a fascinating subject and this video is fantastic.. The presentation by Professor Brailsford is supremely well done. It is hoped this gentleman's students understand what a special opportunity they have had.. Wow.

[@brianmessemer2973]

5:40 "and occasionally, the '41icity' of this is breaking through the murk..." Better than some fiction, this man's ability to tell a factual story. Sounds like space Tolkien.

[@timpierson8371]

Professor Brailsford I wanted to tell you how much I love your lectures. You remind me so much of my grandfather who was a very kind and gentle man and always wanted to spread his knowledge. I learned so much from him as I have with you. Tim

[@nickstubbings]

really enjoying the series, the enthusiasim and knowledge is presented really well!

[@edwardtait4285]

Thank you for your time and wonderful explanation. Best regards.

[@timgluckman8663]

This guy is the best I've seen at explaining BP. Great TV manner

[@jeffbrunton3291]

Thanks, I have read a few accounts on Tunney but this is the best on how the actual machine / encoding worked

[@James_Bowie]

After this very detailed explanation of the delta technique, I now understand how Colossus was used. 👍

[@tocsa120ls]

1:25 I think it is worth mentioning that if somebody wants to listen to this Tunny traffic, they only need a $80 'world receiver' radio, as some German remote weather reporting stations still use it to this day!

[@Roxor128]

I wonder if it's just a software implementation on something like a Raspberry Pi?

[@johnfrancisdoe1563]

Roxor128 The Raspberry Pi is a computer, so it'll be software either way.

[@b43xoit]

Out of what motivation they use it? Why don't they report weather conditions in the clear?

[@davidcottrell1308]

Love this guy...what a FANTASTIC source of real history....truly fascinating!!!!

[@georgehelliar]

I've watched all thsee enigma / colossus videos and I'm completely blown away by the intelligence that was brought to bear on these problems. I'm not a mathematician, but I like to think I'm pretty decent at attacking problems of logic, and even with these beautifully simplified explanations, I'm barely able to get my head around the process. The brainpower required to reverse engineer these ciphers is utterly phenomenal. I've also watched the turing test videos, and learned the professor's opinions on AI, but it seems to me that mastering conversational English would be far easier than what these guys achieved.

[@wktodd]

Thanks Prof:-) I had the pleasure of talking to Tony Small in the early days of the collosus rebuild who was trying to explain this statistical analysis to me but i did not get the importance of the delta method.

[@deviljelly3]

I lived next to Dollis Hill... An amazing building and to imagine that Sir Tommy was there...

[@profdaveb6384]

Sadly he never made it to being "Sir Tommy" (though he should have done) Instead he wa s awarded the lowest "Member" grade of the Britsih Empire Order (MBE) . Indeed I don't think anybody got a knighthood for anything Enigma or Colossus related. Turing was awarded (and accepted) the "Officer" grade of medal (OBE). Max Newman was offered OBE and turned it down as "derisory".

[@stensoft]

Those English pronunciations of Greek letters … until he has shown the letter, I had no idea what he was talking about.

[@JivanPal]

We basically always (particularly at the end of words) pronounce Greek ι (iota) as "eye", /aɪ/, e.g. • π, "pi", is pronounced "p-eye". • ψ, "psi", is pronounced "p-sigh" or "sigh". • ε, "epsilon", is sometimes pronounced as "ep-sigh-lon", especially by the previous generation of mathematicians, as I've noticed. Greek η (eta) is pronounced "ay", /eɪ/, or like the actual Greek as "ee", /iː/, e.g. • β, "beta", is pronounced "bay-taa", or "bee-taa". • θ, "theta", is pronounced "thay-taa" or "thee-taa".

[@bryanl1984]

Yeah... I saw This before that part and expected it to be jarring but, this is exactly the same as American pronunciation... where are you? Also, isnt this guy originally from America but, has lived in the UK for the lion's share of his lufe?

[@stensoft]

I know the pronunciation of pi but the pronunciation of chi and psi really threw me off. I am from the Czech Republic and we (and Germans where I also studied) pronounce the letters in the same way they are pronounced in Greek: psi as psee, chi as khee (with kh pronounced as Spanish j).

[@bryanl1984]

Like a cyrillic Ж / Zhe? I've never seen that in the US. As a matter of fact, that sound doesn't even exist in the Anglosphere.

[@stensoft]

​@@bryanl1984 It's similar to a Cyrillic Х (because it is historically the same letter). Formally, it's an aspirated k. In English, the most common transcription is kh but people usually read that as k'h, like two letters, instead of as an aspirated k.

Thank you Professor Brailsford!

[@jacquesfall8117]

Legalise recreational Mathematics!

[@bryanl1984]

If you have too many numbers on you, are you charged with intent to distribute maths?

[@kevinbyrne4538]

@@bryanl1984 -- I was caught with a random number generator and charged with disorderly conduct.

[@bryanl1984]

@@kevinbyrne4538 Hmmm.... Your story doesn't quite ADD up... If anything, your just being IRRATIONAL! .... I'll let myself out...

[@hanniffydinn6019]

There was a funny red dwarf episode with this premise. Very geeky show.

[@piponope]

Will it give you the muchies?

[@forthrightgambitia1032]

An interesting video would be about the Venona project and cryptanalysis of Soviet messages, which had a role in both the Cambridge 5 saga and the history of the development of the hydrogen bomb (the development itself very important in the history of computing) and Soviet efforts to retrieve information from the Manhattan project - such as how the Rosenbergs and the physicist Klaus Fuchs were caught out.

[@silkworm6861]

Fascinating stories!

[@ianleitch9960]

I love the technology involved. Still a bit beyond me, but I certainly get the idea. I would have loved to have gone to the GC&CS, but that was just before my time. Lowly green key operator ...

[@NoahSpurrier]

Stunning bit of history.

[@pigboiii]

If this man would have been my teacher/lecturer I would have no other option but to pay attention and be facinated

[@ShakalDraconis]

After watching these sorts of videos on Computerphile, I find myself suddenly wondering about the Allies encryption methods, and the German methods to break them. Other then the Navajo Codetalker program, I don't think I've ever heard anything about how message encryption was done on the Allies side.

[@JamesCollins80]

UK had something called Typex which I believe was a copy of an commercial Enigma bought for a business (because who cares about patents in wartime) with some of the security holes filled in by doubling some things up/adding things on. America had their own system as well and there was a combined system built when they joined in later on - all similar rotor machines to this one.

[@anthonyvancampen6729]

@@JamesCollins80 Type X and the American SIGABA were to some extent inspired by the Enigma system but independently developed and debugged. Type X and SIGABA closed the hole created by can't encode a letter as itself. Also, the Type X and SIGABA closed the stutter cycle length issue with Tunny. Lastly, SIGABA and Type X eliminated a major source of errors in messages by creating a message tape and completely avoiding hand transcribing of the enciphered message.

[@davidwilkie9551]

Great stuff.., given an expectation that someone somewhere somewhen has thought of this before, how's the "decryption" of the Standard Model/ universal Enigma code going? (Begs the question) SM phenomena being statistical wave-packages of node-wheels, or "common denominators" hypo-frequency/"below" the ground state positioning boundary of flat Spacetime, composed of machine memory/time-duration, ..dominant primes and co-prime factors of "leaky brane"+/- => annealed, ..phase-locked multi-phase state resonance... (Math-Phys-Chem)

[@AndersJackson]

I like that he talked about the original meaning of Computer. :-)

[@cupajoe99]

A lot went over my head, but still super interesting.

[@hanniffydinn6019]

Learning about Bletchley is an endless joy of computer science for me. I'm still horrified I wasn't taught computer history at uni. The British government still tries to keep it all secret.

[@parrotraiser6541]

About "computer" meaning a person; In the very early days, (19th century) of office machinery, "a typewriter" meant the user, not the machiner. In one of G.K. Chesterton's books, (I think it was "The Man who Was Thursday") he refers to a man "falling in love with a typewriter", which confused me no end when I read it. (The book was supposed to be a bit surreal, but that seemed to be taking kinky and weird too far.)

[@simondodge9733]

I have an ancestor in a census record whose Occupation is Computer. Always makes folks laugh.

[@jeffbrunton3291]

The interesting point for me about Tunney is that despite having Colossus later on, it always required codebreakers to identify part of the pattern

[@sketin]

This video doesn't make me sigh at all, deep or otherwise. :)

[@piccalillipit9211]

THAT was amazing

[@MrDaiseymay]

Fantastic ! everything crystal Clear now ?

[@alexandermayr7]

love that shirt!

[@Neumah]

This makes me think of Arne Beurling who cracked the Siemens and Halske T52 in two weeks.

[@MartinTowell]

Love these videos about the decryption of the messages but what about the history of the encryption side of it? Who came up with the idea of using those sets of wheels? Why did they decide to use xor instead of some other form of encryption?

[@dustysparks]

XOR was basically THE form of encryption when it came to teletype printers. It had been "invented" I think only a decade or two before, so it was still cutting edge, and remember there was no such thing as a "memory" in those machines, it was either on paper tape (and thus fixed, and usually relegated to typed plaintext for transmission or ciphertext taken from the airwaves), or mechanical switches set only at the start of a run, or it was the patterns inside wheels on a rotary device of some kind, which were all the rage then (Enigma probably the most famous but only one of dozens in use at that time). Addition and subtraction introduced complexities (carries and borrows) that XOR simply didn't need or have, so it was also far simpler to implement. All the oops's and gotcha's and perils of XOR were discovered as a RESULT of its heavy usage in all these machines. XOR is still a huge part of encryption today, AES-GCM which is used in all modern browsers is AES in a counter mode (the input to the cipher in Galios Counter Mode is a large binary number, not the plaintext, and the output is XOR'd with the stream of plaintext to make the cipertext). The bug in WPA that was discovered a while back (and yes there's a Computerphile video for that as well) is basically forcing a "depth" in AES-GCM and using that to get the key-stream for the session without ever having to break the original session key itself.

[@profdaveb6384]

Yes, XOR encryption was proposed by Gilbert Vernam and patented by Bell Labs in a form that required the key stream to be provided on a tape that ran in parallel to the plaintext.. The idea of creating a bolt-on second stage to a conventional teleprinter that could electro-mechanically provide a pseudo-random key stream was extremely attractive. Such a machine could be handled by one operator and had an 'operator friendly' keyboard very much in the spirit of an ordinary typewriter. These advantages tended to sweep aside objections from some mathematicians that additive ciphers, such as XOR, should be totally ruled out for cryptographic use because of their appalling vulnerability to "depths".

[@b43xoit]

XOR provides symmetric ciphers, which in turn provide more convenience in operation than asymmetric ones.

[@bernardo013]

What happens when the intercepted messages have typos? Was this a problem that happened often, and if so how did they get around it?

[@tocsa120ls]

Germans and _mistakes_? Have you been drinking? Those messages were PERFECT!

[@johnfrancisdoe1563]

bernardo013 Typos would go in the same statistical buckets as unknown words. They would become readable after finding the Key and Perturbation settings that completely break the encryption. However the delta technique would work equally well on typpos and tyypos as it did on happy.

[@deviljelly3]

Wonderful

[@BuickDoc]

I recall back in 1965 my introduction to computers on a CDC 6600. Of course, I didn't get to touch it, but I could see it through a plate glass window. I seem to remember spending an inordinate amount of time transposing base 10 numbers by hand to binary, octal, and hexadecimal. Today I understand why binary, but why octal and hexadecimal? I can't remember...Thanks!

[@anthonyvancampen6729]

Many computers like the Burroughs machine that became the US Navy's Mk 152 used an operator display and control panel that grouped the lamps and switches in groups of three (octal). In fact, to this day US Navy aircraft use side numbers that are coded in Octal. The take away for the use of both octal and hexadecimal is that they are mostly just a more compact notation scheme for binary.

[@tildessmoo]

I imagine Tutte was counting the encryption steps backwards, since the process of decryption is the reverse of encryption. So, using the Latin alphabet, it would go last through the 'z' wheel, before that through the set of 'y' wheels, and before that through the 'x' wheels. He just used the Greek alphabet instead, so χ, ψ, ω are the last three letters. As to why he called the last wheels μ instead of ω, I have no idea.

[@nicolabaccillieri6503]

Hi Computerphile, I love your video. Excellent explanations! Can you make a video about AES 128/256/512 and Bcrypt? Thank you, keep it up! ;)

[@hennobrandsma4755]

AES 512 does not exist. Bcrypt is not a cipher algorithm.

[@nicolabaccillieri6503]

@@hennobrandsma4755 yeah, Sorry, 128-192-256. I know bcrypt is not a cypher algorith, but its interesting how It works

[@forthrightgambitia1032]

Surely chronologically dealing with DES would make sense first of all.

[@evaluati0n_]

just in time to distract me from my java and computer systems assignment

[@dontaskiwasbored2008]

So cool.

[@ThisRandomUsername]

At the end of the video there's no way to get to the 2 highlighted videos. Is it meant to be a clickable box? KZbin has taken away annotations for videos.

[@James_Bowie]

Wide 'eyeline' fan-fold printer paper. Haven't seen that for decades.

[@ForbinKid]

I wonder if the 1970 movie Colossus the Forbin Project had any ties with this.

[@adisonwail6522]

Is there a book on stuff like this like encryption machines and encryption?

[@black_platypus]

17:02 "Navel Enigma"? "Yes, my belly button is very complicated" :P

[@RenX3133]

5:24 "But hexedentally ..." :3

[@desmondbrown5508]

It makes me wonder how many of these security flaws were intentional. We know there were some Germans that didn't like what was happening and so they either sabotaged war machines on purpose or at least altered them to be less effective. I'm curious if the initial failures were intentional, knowing that if someone figured out the simpler problems that the fixes would then also be able to be decrypted as well.

[@auto_ego]

In this video, we see what happens when no one interrupts Prof. Brailsford.

[@StreuB1]

PROF!!!!!

[@greyed]

The 2nd set of wheels, the Quake wheels... Admit it, you thought of it too when he wrote it down.

[@TheTurnipKing]

computer: An automaton for computing

[@jeffbrunton3291]

There is a memorial to Tuttle in Newmarket with message - in code....

[@aaaaaaaaaaa5820]

Did this whole video exist so that he could make that sigh pun

[@lauriepocock3066]

I assumed BT referred to Chi wheels because that was the significance test he used.

[@kisame_5331]

Who disliked this video ? This is so interesting

[@HebaruSan]

How were the allies securing their own communications at the time?

[@hennobrandsma4755]

SIGABA (US), Typex (UK), and some other machines. SOE used double transpositions and one-time pads. In the early war years the navy used older code books systems which were broken by Japanese and Germans.

[@U014B]

What if χ and ψ were the encyphered forms of α and β?

[@anthonyvancampen6729]

LOL

[@jimbrowsing5690]

If the Germans had removed the double letters from the messages, would an even bigger/cleverer machine/system have been devised to break the code ?

[@spoonikle]

Weird. I could have sworn I subscribed the very first video ever posted on the channel... Just found out I was unsubscribed.. perhaps for weeks.

[@alexandterfst6532]

Why no subtitles !!!

[@daviddupoise6443]

@24:15 - Yeah "it's all perfectly straightforward", Ha yeah in a very interferency sort of way. Of course it is, of course it is. ;)

[@Trav2016]

Can we get American English subtitles,? It's hard to get the nuances sometimes of the lessons and history being described.

[@profdaveb6384]

I have them under way but 37 mins takes some doing .... :-)

[@garyslatter9854]

Fabulous instate into early #GCHQ and #NSA origins...

[@garyslatter9854]

incite

[@jeffirwin7862]

[deep sigh]

[@Jivvi]

"Digital computer" could just mean someone who counted on their fingers.

[@donaldasayers]

Just noticed we've got the same curtains.

[@TheSharkey22]

Donald Sayers I have the same curtains......

[@TheSharkey22]

....as his shirt.

[@brianmessemer2973]

Geez, I understand where you're coming from. I want to have something in common with this incredible man too - I don't care WHAT it is!!! 😂👍

[@peppers1758]

Why didn't they just use 128 bit encryption

[@hanniffydinn6019]

Amen break, is the greatest break !!!

[@MrDaiseymay]

AND TILTMAN.

[@billsheehy1]

After watching this guy, I realized that I am actually stupider than I thought.

[@danmimis4576]

paused at 4:02 to ask this before I forget: lemme understand something here: when you encrypt a word with a key not only that you don't have to just simple add them (in binary or whatever), but there is an infinite number of ways of doing it and and not even a quantum computer can deal with the nasty infinite, right? Or am I missing something?

[@garfieldfarkle]

Only if the key is perfectly random and never repeats. This is what one-time pads are supposed to do.

[@b43xoit]

Large, but finite.

[@LarryTheRoleplayerTM]

This is why computers are so hard