WEBVTT 00:01.010 --> 00:09.370 Of any episode that I do in this entire series this is my favorite because this my friends is cryptography. 00:09.440 --> 00:17.600 Cryptography is the science the study of taking data and making it hidden in some way so that other 00:17.600 --> 00:19.360 people can't see it. 00:19.460 --> 00:21.270 And then bringing the data back. 00:21.320 --> 00:27.200 So the big word we want to use when we talk about cryptography is taking some kind of information and 00:27.200 --> 00:30.130 providing confidentiality to that. 00:30.140 --> 00:35.600 Now we do this through a number of different ways but the magic word I want to use right now is obvious 00:35.600 --> 00:43.400 occasion obvious occasion is to take something that looks like it makes sense and to hide it so that 00:43.400 --> 00:47.260 it does not make sense to the casual outside observer. 00:47.630 --> 00:53.210 Now theres a lot of different ways we can do obfuscation to provide confidentiality What are the things 00:53.210 --> 00:54.710 we can do is diffusion. 00:54.710 --> 00:57.740 So here is a picture of my grandson right here. 00:57.740 --> 01:00.320 Let's take a look at that that Steve and then the gorgeous. 01:00.320 --> 01:05.870 So what I'm going to do is I'm going to diffuse this image and make it fuzzier. 01:05.870 --> 01:10.460 Now in this particular case we didn't diffuse it too much so if you look at it you could probably tell 01:10.460 --> 01:13.560 there's still a cute little kid there and he is cute. 01:13.940 --> 01:22.220 But diffusion only allows us to make it less visible less obvious. 01:22.250 --> 01:27.290 Now what we can also do is confusion so let's take that same image of Steve one more time just stir 01:27.290 --> 01:27.550 it up. 01:27.560 --> 01:29.340 Let's just make a mess out of it. 01:29.660 --> 01:34.370 Now in this particular case we've created a lot of confusion of the image and it basically stirred up 01:34.760 --> 01:40.400 and it would be very difficult for somebody to simply look at this and go oh that must be Mike's grandson 01:40.400 --> 01:41.220 Stephen. 01:41.240 --> 01:46.670 So the other challenge we have with cryptography is we go through this process of taking some kind of 01:46.670 --> 01:50.100 data and we go through this obfuscation process. 01:50.240 --> 01:57.350 But and here's the important one is that we've then got to take this obfuscated data and some way bring 01:57.350 --> 01:59.880 it back into its original form. 01:59.960 --> 02:06.910 So we call this encryption and decryption and cryptography is the process of making this happen. 02:06.920 --> 02:09.530 Cryptography has been around for a long long time. 02:09.800 --> 02:16.040 In fact probably one of the oldest type of cryptography has ever been around is something called the 02:16.050 --> 02:20.040 Sezer cipher. 02:20.120 --> 02:25.130 I don't know about you but when I was a little kid and I was eating a box of cereal you dump out the 02:25.130 --> 02:28.600 box of cereal and you'd get some kind of prize inside. 02:28.600 --> 02:31.250 And one of the things we'd get is called a secret decoder ring. 02:31.280 --> 02:32.290 So what I want to do here. 02:32.290 --> 02:34.680 Let's can we put up a picture of a secret decoder ring real Correct. 02:34.700 --> 02:34.990 OK. 02:35.000 --> 02:38.210 So this is a classic old school kids decoder ring. 02:38.210 --> 02:42.620 Now what I want to do is I've made my own decoder ring right here. 02:42.800 --> 02:45.330 And I'd like us to take a little peek at this guy. 02:45.740 --> 02:49.880 And what you're going to see here is I basically got a wheel with all the letters of the alphabet A 02:49.880 --> 02:51.890 through Z and on the inside. 02:51.920 --> 02:54.250 I've got letters the alphabet aither Z. 02:54.290 --> 03:00.580 So right now I've got them lined up eight a b to b c to see now what I can do to make a secret code. 03:00.740 --> 03:04.450 Let's say I can turn this to I can rotate this two times. 03:04.650 --> 03:11.810 And what we can do is we can take our original plane information what we call the plain text message 03:12.230 --> 03:13.470 that we want to encrypt. 03:13.670 --> 03:18.220 And we just change the letters so we call this substitution. 03:18.230 --> 03:21.530 What we'll do is we'll take one value and substitute for another. 03:21.530 --> 03:23.920 Now in this case I've rotated it twice. 03:24.020 --> 03:27.730 So we actually have a term for this we call it r o t to. 03:27.770 --> 03:28.810 Just like that. 03:29.120 --> 03:32.710 And if I turned it three times it would be Arati 3. 03:32.750 --> 03:38.360 Now so we can take like the word ace A C E. 03:38.750 --> 03:44.560 And I can change a c e to see. 03:46.140 --> 03:47.740 G get the idea. 03:47.850 --> 03:51.620 So that's the cornerstone of the Sezer cipher. 03:51.790 --> 03:58.200 So to convert something with a Caesar cipher and perfect that just go through the process real quick. 03:58.200 --> 04:03.550 Now what I'm going to do let's put up a piece of plain text that we want to encrypt. 04:03.600 --> 04:05.210 We attack at dawn. 04:05.220 --> 04:06.900 So here it is we attack at dawn now. 04:06.960 --> 04:09.610 First we are going to do get rid of all the spaces. 04:09.630 --> 04:12.240 So now it just says we attack at dawn very readable. 04:12.270 --> 04:15.750 We don't worry about upper or lower case in this particular situation. 04:15.760 --> 04:23.320 Now let's go ahead and put our groovy little secret decoder ring up there and let's go ahead and turn 04:23.320 --> 04:26.380 it in this particular case five times. 04:26.380 --> 04:26.960 All right. 04:27.040 --> 04:28.970 So we're doing an OT 5. 04:28.970 --> 04:34.050 So let's turn it one two three four five times. 04:34.070 --> 04:39.910 So now what we can do is by using the secret decoder ring we can go ahead and encrypt. 04:39.970 --> 04:42.330 We attack at dawn as follows. 04:43.340 --> 04:47.600 So what we now have generated is a classic Caesar cipher. 04:47.720 --> 04:51.380 Now there's a problem with Sezer ciphers and the biggest problem we have with the more Anything else 04:51.410 --> 04:57.920 is that in fact people who like crossword puzzle books will pay money to be able to do this is that 04:58.100 --> 05:00.870 we can decrypt them. 05:00.920 --> 05:06.020 We can just by looking at them we can provide what's known as cryptanalysis. 05:06.020 --> 05:10.370 Cryptanalysis is breaking these encrypted codes. 05:10.550 --> 05:17.850 So the problem is is that anybody who's good at these things could pretty easily crack this so the Caesar 05:17.870 --> 05:23.570 cipher even though it is a substitution cypher does have the problem is that it's just too easy to predict 05:24.020 --> 05:26.600 what this is because we're used to looking at words. 05:26.870 --> 05:29.870 So what I want to do is make it a little bit more challenging. 05:29.870 --> 05:35.660 The first thing I'd like to try to do is I want to bring in something called the V-J cipher 05:38.400 --> 05:44.400 the cornerstone of vision a cipher is that it's really just a Caesar cipher with a little bit of extra 05:44.790 --> 05:46.170 confusion involved. 05:46.170 --> 05:52.800 So what I've got here is a table that shows all the possible Sezer ciphers there are. 05:52.800 --> 05:56.010 So here at the top we're going to have the word plain text I'll show you how that works. 05:56.020 --> 06:02.220 Just a minute and then you'll see on the far left hand side it says 0 through 25. 06:02.220 --> 06:08.670 So these are all the possible Arati values you can have from our zero which means a equals a big will 06:08.730 --> 06:12.480 be all the way down to our OT 25. 06:12.480 --> 06:16.980 So what we're going to do let's hold onto this for a minute and now let's go ahead and start with a 06:16.980 --> 06:21.210 piece of plaintexts let's use we attack at dawn one more time. 06:21.420 --> 06:25.250 And what I'm going to do this time is we're going to apply a key. 06:25.290 --> 06:28.560 The key is simply a word that's going to help us do this encryption. 06:28.650 --> 06:34.150 In this particular case I'm going to use the word face f a c e. 06:34.590 --> 06:42.330 Now what I'm going to do is I'm going to put F A C above the first four letters of we attack at dawn. 06:42.330 --> 06:47.810 So here we go F A C and then I'm going to just keep repeating that so it put face again. 06:48.030 --> 06:51.240 I'm going to put a face again and you'll see I've got two letters left over. 06:51.240 --> 06:51.830 No big deal. 06:51.830 --> 06:53.280 I just put a. 06:53.550 --> 06:57.930 And what we've done is we have applied a key to our plain text. 06:57.930 --> 07:05.760 Now what we're going to do is we're going to use the key to change the Caesar cipher OT value for every 07:05.760 --> 07:06.920 single letter. 07:06.930 --> 07:08.730 So let's go ahead and do this. 07:08.820 --> 07:11.190 Now what I'm going to do is I'm going to reference my little chart here. 07:11.190 --> 07:15.550 So the first letter of the plaintext is the W in we. 07:15.600 --> 07:18.150 So here's the W up at the top. 07:19.250 --> 07:23.100 And the key value is f.. 07:23.130 --> 07:27.250 So let's go down on the y axis here till we get to an F.. 07:27.330 --> 07:29.610 They see that f you'll see the number five right next to it. 07:29.610 --> 07:31.580 So this is Aro T5. 07:31.590 --> 07:38.620 So all I need to do is do the intersection of these and we get the letter B Great let's do it again. 07:38.650 --> 07:42.100 Now in this case the second time it's the letter E from we. 07:42.400 --> 07:48.520 And in this particular case the key value is a which is kind of interesting because that's our goatees 07:48.880 --> 07:49.360 zero. 07:49.360 --> 07:50.690 But that still works. 07:50.710 --> 07:53.000 So we start up at the top find the letter e. 07:53.440 --> 07:54.530 Then we find that a. 07:54.580 --> 07:56.790 And in this case because it's our OT 0. 07:56.830 --> 08:01.320 By coincidence e is going to stay as e. 08:01.450 --> 08:04.990 Let's do the third value now this time it's the A's in attack. 08:05.110 --> 08:10.330 So we go up the top there's the letter A and the key value is C is Charlie. 08:10.360 --> 08:11.860 So we go down to the C. 08:11.950 --> 08:19.150 That's our T2 and we then see that the letter A is now going to b c. 08:19.310 --> 08:21.080 Ok let's do one more real quick. 08:21.190 --> 08:24.600 In this particular case it's the first T and attack. 08:24.700 --> 08:31.200 So we come over to the T's and now the key value is e is in is in phase so we go down here. 08:31.210 --> 08:34.120 That's our code for we do the intersection. 08:34.120 --> 08:36.500 And now we've got an X.. 08:36.520 --> 08:42.460 So the first four letters of our encrypted code is B E C X.. 08:42.460 --> 08:45.600 So let's go ahead and run through and do the rest of these real quick. 08:45.610 --> 08:52.690 Just put those in for you and we have now encrypted in the vision a style. 08:52.730 --> 09:00.080 So the beauty of the day is that it actually gives us all the pieces we need to create a classic piece 09:00.080 --> 09:01.310 of cryptography. 09:01.310 --> 09:07.790 Number one we have an algorithm and the algorithm are the different types of Sezer ciphers and the rotations. 09:07.790 --> 09:15.700 And second we have a key and the key allows us to make any type of changes we want within our O.T. 0 09:15.710 --> 09:19.930 to our code 25 to be able to encrypt our values. 09:20.240 --> 09:25.070 Any algorithm out there is going to use a key in today's world. 09:25.070 --> 09:29.870 So when we're talking about cryptography today we're always going to be talking about algorithms and 09:29.870 --> 09:30.830 keys. 09:30.830 --> 09:36.180 Now the problem with the vision day is that well number one it's surprisingly crackable Isn't that interesting. 09:36.260 --> 09:41.240 But the bigger problem is his vision a works just great for letters of the alphabet. 09:41.480 --> 09:52.220 Unfortunately it's terrible for encrypting pictures or sequel databases or your credit card information 09:52.700 --> 09:53.920 in the computer world. 09:53.930 --> 09:57.660 Everything is binary everything is ones and zeros. 09:57.680 --> 10:05.480 So what we need to do is come up with algorithms that provide the type of encryptions and descriptions 10:05.480 --> 10:10.550 we need to encrypt and decrypt long strings of just ones and zeros. 10:10.550 --> 10:13.970 Now if you look at a string of ones and zeros you go hot. 10:14.000 --> 10:17.210 How does anybody make anything out of this will they do that. 10:17.230 --> 10:21.920 There's a you've got a string of ones and zeros may look like nothing Utu a human being but to Microsoft 10:22.100 --> 10:28.310 Word that could be a word document or it could be a voice over IP conversation or it could be a database 10:28.310 --> 10:29.500 stored on a hard drive. 10:29.540 --> 10:30.340 I don't know. 10:30.350 --> 10:35.210 But just because as human beings we look at long strings of ones and zeros and get confused. 10:35.210 --> 10:37.680 Trust me the computers don't. 10:37.840 --> 10:43.450 What we need to do though is we need to come up with algorithms which unlike Caesar's or vision visionaries 10:44.050 --> 10:46.270 that will work with binary data. 10:46.300 --> 10:50.100 Now luckily for us there are a lot of different ways to do this. 10:50.110 --> 10:57.550 So what we're going to do is I'm got an example of a algorithm we're going to use that encrypts a simple 10:57.550 --> 11:03.040 phrase that we're going to we're going to convert those to binary by the way and we're going to do this 11:03.400 --> 11:10.870 using a very interesting type of binary calculation called exclusive or for our first encryption. 11:10.870 --> 11:12.960 I'm going to encrypt my name. 11:13.000 --> 11:16.260 So here's my name Mike and I K E. 11:16.450 --> 11:21.550 Now the first thing we have to do is we're going to have to convert this to the binary that a computer 11:21.550 --> 11:22.370 would use. 11:22.390 --> 11:28.060 So I'm going to be using the binary equivalence of these text values. 11:28.180 --> 11:31.600 So let me go ahead and convert these into their binary equivalents. 11:31.930 --> 11:37.750 And anybody who's ever looked at ASCII code or Unicode should be aware that we can convert these into 11:37.750 --> 11:38.190 binary. 11:38.230 --> 11:39.370 OK so here we go. 11:39.370 --> 11:42.630 So there's n i k e. 11:42.850 --> 11:44.730 Converted into binary. 11:44.740 --> 11:53.110 Now notice that each character takes eight binary digits so we got 32 bits of data that we need to encrypt 11:53.110 --> 11:54.670 So that's our clear text. 11:54.670 --> 11:59.650 Now in order to do this we're going to need two things First of all we're gonna need an algorithm and 11:59.650 --> 12:01.270 then we're going to need a key. 12:01.270 --> 12:06.250 Now keep in mind what I'm making up is like the simple version of encryption you can possibly do. 12:06.250 --> 12:06.840 So let's go ahead. 12:06.850 --> 12:09.040 First of all set up our algorithm. 12:09.040 --> 12:14.620 Now our algorithm is extremely simple using what we call an exclusive or so. 12:15.340 --> 12:18.220 Here's the exclusive or this is what we call a truth table. 12:18.220 --> 12:23.680 So what I'm going to do is I'm going to choose because this Mike algorithm arbitrarily chooses this 12:24.010 --> 12:27.870 is I'm going to be using a five bit key. 12:27.880 --> 12:32.090 Now there's a reason I'm using a tiny tiny shortcut like this in the real world. 12:32.110 --> 12:37.180 Keys can be thousands of bytes long but for right now we're just going to use a five bit key. 12:37.180 --> 12:40.390 So to make this work let's start placing the key. 12:40.390 --> 12:42.970 So I'm going to put the key over the first five bits. 12:42.970 --> 12:49.420 So here at the letter M for Mike and now we can look at this table and we can start doing the conversion. 12:49.420 --> 12:57.220 So let's convert those first two values then the next then the next then the next. 12:57.220 --> 12:59.550 OK so now we've converted a whole piece worth. 12:59.560 --> 13:04.850 But in order to keep going all we have to do is shlepped that key right back up there OK. 13:04.870 --> 13:08.120 In fact let's go ahead and extend the key all the way out. 13:08.470 --> 13:14.290 So now the key Do we just keep repeating it and you'll see here at the end it doesn't quite line up. 13:14.380 --> 13:15.400 No problem. 13:15.400 --> 13:20.060 Just add whatever amount of key you need to go ahead and fill up the rest of this. 13:20.060 --> 13:22.140 So there we go. 13:22.390 --> 13:25.160 OK so now we can go ahead and complete this. 13:25.180 --> 13:26.810 Now let's just do it fast. 13:26.810 --> 13:33.050 You can double check me if you want but as we go through here using the exclusive or algorithm we then 13:33.050 --> 13:39.920 create our ciphertext go and spot check a couple of those for me and make sure I got them right. 13:40.930 --> 13:41.490 OK. 13:41.600 --> 13:42.580 Beautiful. 13:42.680 --> 13:45.830 Now so this is the cipher text. 13:45.830 --> 13:48.500 Notice that we have an algorithm which is extremely simplistic. 13:48.500 --> 13:50.800 We have a key which is very very simple and short. 13:50.990 --> 13:58.090 But we now have an absolutely perfect example of binary encryption. 13:58.160 --> 14:02.720 Now to decrypt this we'd simply reverse the process so I'm not going to go through all that but appreciate 14:02.990 --> 14:09.020 that we would take the ciphertext play the place to keep to it and then basically run the algorithm 14:09.020 --> 14:14.780 backwards and then we would have the decrypted data. 14:14.790 --> 14:20.970 So even though we've only seen three types of encryption so far keep in mind that these really do everything 14:20.970 --> 14:23.700 that we're going to be seeing in later episodes. 14:23.730 --> 14:25.840 Just a lot more complicated now. 14:25.890 --> 14:27.420 A couple of things we need to think about here. 14:27.420 --> 14:33.180 First of all what's interesting is that if we always have an algorithm and a key there is a gentleman 14:33.270 --> 14:40.230 named August Kirchoff who came up with a very very interesting concept Kirchhoff's principle says this 14:41.160 --> 14:45.540 As long as you don't know what the key is to an encryption. 14:45.660 --> 14:48.750 You can actually understand the algorithm completely. 14:48.750 --> 14:50.990 Now this is really really important. 14:51.030 --> 14:58.050 Today's big super duper encryption tools that we use out there to protect you on the internet are all 14:58.110 --> 14:59.200 open standards. 14:59.220 --> 15:02.270 Everybody knows how the algorithms work. 15:02.280 --> 15:03.350 Now you would think wait a minute. 15:03.360 --> 15:08.190 Now if I know how the lock works in essence wouldn't I be able to pick it easier. 15:08.220 --> 15:11.710 And the answer is interestingly enough no. 15:11.760 --> 15:19.400 In our society by showing everybody the walk everybody can check the lock to make sure it isn't pickle. 15:19.410 --> 15:25.740 So when we talk about proprietary encryption everybody gets nervous because if we don't all know how 15:25.740 --> 15:29.290 the lock works we can't all test the lock to make it work. 15:29.310 --> 15:35.580 So Kirchhoff's principle and something we stand to today simply says everybody knows the algorithm. 15:35.580 --> 15:38.620 But if you don't know the key it's not going to do you any good. 15:39.530 --> 15:44.870 So last thing I want to talk about in cryptography basics is the idea of what data are we encrypting. 15:44.870 --> 15:48.230 Because in the world data is all over the place. 15:48.530 --> 15:54.350 The first place we might think about is what we call data at rest now data at rest simply means something 15:54.350 --> 15:59.380 stored on a hard drive or on optical media or on a thumb drive or whatever it might be. 15:59.630 --> 16:05.540 And in these types of cases we encrypt the data when we put it onto the drive itself so it sits on the 16:05.540 --> 16:08.520 stored media in encrypted format. 16:08.570 --> 16:11.180 The other one is called data in transit. 16:11.180 --> 16:19.610 So if I've got a voice over IP call or I'm sending a text message that data is moving it's going through 16:19.610 --> 16:22.130 the internets and all the different tubes. 16:22.130 --> 16:26.760 And is that data to be encrypted while the data is in transit or not. 16:26.780 --> 16:27.960 This is a big issue. 16:28.400 --> 16:31.160 And the last one is data in process. 16:31.250 --> 16:36.950 So we take a big database and we pull it off of a hard drive when we start calculating on that database. 16:36.950 --> 16:40.210 So it's sitting in RAM or its citizens CPQ. 16:40.370 --> 16:46.250 Those are important areas for us to consider when we're talking about cryptography. 16:46.250 --> 16:57.610 Where are we going to be encrypting and decrypting that data. 17:01.540 --> 17:07.930 In.