Pragmatic 47: Interrogation Signal

Welcome to Pragmatic. Pragmatic is a weekly discussion show contemplating the practical application of technology. Exploring the real-world trade-offs, we look at how great ideas are transformed into products and services that can change our lives. Nothing is as simple as it seems. This episode is sponsored by Audible. Please visit audiblepodcast.com/pragmatic for a free audiobook download today. This episode is also sponsored by Mandrill. Mandrill is a transactional email service that easily ties into your website and apps when you need to send one-off emails like responses, password resets, acknowledgments, and so on. Visit mandrill.com and sign up today. Why not? It's free. And use the promo code "Pragmatic" to get 50,000 free email transactions per month for 6 months. Normally there's only 12,000 a month. That's four times the normal amount. Integrate, deliver, track and analyze using InMile infrastructure from Mandrel. We'll talk about them more during the show. I'm your host John Gigi and today I'm joined once again by my co-host Vic Hudson. How you doing Vic? I'm good John, how are you? I am very very good. Is it a bit cold over there? It's not quite so cold right now. It was much colder last weekend but it's chilly. Chilly vanilla. There you go. Well, it's actually stinking hot over here. So there you go. I'll swap you. I like the cold bill and I like the heat Yeah, we had a record low last week. So now we had it. We have an annoying heat wave last week, too So yeah, I know go figure Anyway, I'd make some and far dismissive passing comment about global warming, but yeah, the Individual a day by day data points do not equal a global trend. So we'll dispense with that Okay, I just want to talk quickly about reviews in iTunes before we get stuck into the episode. So, the show passed another milestone of sorts this week because Pragmatic on iTunes got its 50th written review. Given the show is about a year old now, that's roughly one written review every week, which is awesome. That's pretty cool. It is very cool and to each and every person that's reviewed the show on iTunes, many, many thanks. And I sincerely mean that. If you haven't rated or reviewed the show yet and you'd like to say thank you or tell me how I could improve the show, please leave a review or a rating in iTunes. I mean, another idea, if you're using Overcast, you can recommend your favorite episodes through the app. You just tap the little I information icon for the episode and then tap recommend. It's that simple. Now, I tend to go on cycles when I'm asking for reviews for the show. So this is gonna be my last one for a while because I don't like to ask all the time. But anyway, sometimes I've had feedback that it's wrong to ask for reviews. Honestly, if you want to write a one-star review, go for it. I can't stop you and I wouldn't want to stop you. I mean, tell me what I'm doing wrong. That's totally fine. Yeah, 'cause I mean, obviously, if I'm gonna ask for feedback, then, you know, and anyone-- - You should expect it. - Exactly, I mean, you ask for feedback, you're gonna get it. Yeah, if you don't wanna feed people's feedback, then, you know, quote unquote, don't tell me. But, you know, that's not what I'm saying. I want to know. And, you know, so I'm asking for feedback and I'm probably going to get it. And, you know, and in my mind, that makes it OK to ask because it's a fair and balanced question. So so that's all it's so it's all good is all I'm saying. So leave your feedback positive, negative, neutral. It's all good. And I guess the other thing is, keep in mind, look, I don't I don't do the whole donation thing, right? Yes, we have sponsors, but I don't take donations and stuff. So if you enjoy listening, really, that's enough for me. Seriously, really it is. But, you know, and if you don't want to listen, you don't have to. But if you do enjoy the show, think of rating it or recommending it as a donation of a few moments of your time. And that's all, no dollar cost, just that. And it can really help the visibility of the show. Plus, it lets me know what I'm doing right or wrong and how I can improve. So there you go. That's it. No more sales pitch. We're done. We'll-- and by that we don't mean we've done the show because now we'll get stuck into it. Drops the mic and leaves. I'm not dropping the mic, Vic. But that's mainly because I'm not holding the mic. The boom arm is holding the mic. And the boom arm seems to be quite robust. It's not going to drop the mic on its own of its own accord. And there's no quick release either. So that's not going to happen either. So I'm sorry. Anyway, right. What were we talking about? Oh, yeah, that's right. The topic. This particular one is a request that was-- it was a topic suggestion. So this is not just voted for, this was suggested in the topics list. It was suggested on the 26th of October. So it was only, well, pretty well 4 weeks ago. So 4 weeks ago, Florian Heiber, he's a fan of the show from Germany and he's been a fan of the show since I think almost the very beginning. So he has suggested this topic. And what is the topic tonight? All about product identification, barcodes. That's right. But not just barcodes. Nothing but a number. Nothing but a number. Yeah, I guess. But not just that, I also want to talk about the more modern twist on it, which is RFID. Okay. So, I think it's a great topic. It's one that I never considered, I never thought of, but it actually is one of those topics that really does interest me. I find it fascinating and I look at the way things have progressed. And it is definitely worth talking about and I find it fascinating. So, here we go. Indexing and index codes. I mean the point, okay, got to start with this. The point of indexing data is about finding something later as quickly as possible. That could be anything from a book, a file, a record, piece of information about any of those sorts of things. Indexing allows us to find things when we need them as quickly and as efficiently as possible. And the first one that springs to mind when I think of an indexing system that's in mass use is the Dewey Decimal System. and that was something that a guy called Melville Dewey came up with in 1876 and it's used in 200,000 libraries in over 130 countries around the world but that's an example of an indexing system that's used to assist you in locating books but obviously, and that indexing system is built on numbers I mean, indexes can be represented in lots of different ways obviously numerically, like the duodecimal system, alphabetically, symbolically, you know, it depends on who or what is going to be reading them. So for average people, we'd rather see letters and numbers because that's what we know. We know the alphabet or, you know, if certain, depending on the country that you grow up in, if you're from Asia or Japan or China or, you know, then they've got different lettering systems and numbering systems that's different from Western culture, but it still boils down to the same thing. It's a combination of letters and numbers. You could do it with symbols as well. But honestly, when it comes to technology, that's a problem because technology and its entry level and its ability to interpret numbers, letters and symbols, it's nowhere near as good as our brains for whatever reason. I do sometimes wonder though, because we're about to talk about barcodes, whether or not if people were taught barcodes from birth, if they would be able to recognize them visually, but I don't think so. So, for a whole bunch of different reasons, it's technologically easier for you to read, for a computer to read something in a different format that is not an actual number, a written number, or a letter, or a symbol, exactly. So, it's far better if it is using uniform squares, uniform lines, uniform spacings, all of that stuff. So, that uniformity lends itself to, well, the idea of a barcode. And the thing is that a barcode, if you think about it, previously on episode 39, we talked a little bit about Morse code. In fact, I think I've talked about it on previous episode, even before that. So Morse code, of course, being invented by Samuel Morse and you know, it's... I'm all kind of rubbish, right? Okay, so it was invented for the telegraph because you're transmitting information over long distances and it was the only way because telephones hadn't been invented at that point. But then they started using it because it was really good for low power transmission and it was used in the Navy for a very long time. So, honestly, think about Morse code and what it looks like, dot dash, dot dash, dash, dot dash, whatever, and then spacing between the letters. If you take those dots and dashes and you sort of stretch them out vertically and you create a bunch of lines, hey, Presto, you just created a barcode. And it's actually true that when they were developing barcodes originally, they were inspired by Morse code. So kind of interesting little crossover there. Yeah, that is interesting. Okay, so machines of course, I guess the idea is that you want to, rather than taking a photo of an object and having the machine translate into something usable, you know, what we do instead is we will reflect light, we'll project light off onto a barcode and then we'll read that data off. off. And then what that does is the scanner will then take that information, it'll convert it from a varying signal of reflected light into a bunch of widths and spaces and then it'll run it through a decoder that says "Okay, well I've got this little width here and that gap there and this little width here, I know that's a 1." Okay, and so on and so forth through all the numbers to figure out "Okay, well the number is like 1, 2, 3, 4, 4, 2, 422121111, whatever. Okay. So the how it does that is very, very interesting. So we're gonna explore some of that. So the codes themselves are just numbers, right? They're just a barcode is essentially a visual representation of numbers in a different format that's easier for a scanner to read. That's all it is. So you'll see at the bottom of a barcode, you're gonna see the numbers written. Those numbers aren't actually read by the scanner. They're just there for the humans. So that if you try and scan the barcode and it doesn't work, you can then punch the numbers in manually. And that's just how it works. I mean, you've worked at point of sale. I've worked at point of sale. I'm sure that you've had a barcode that hasn't worked at some point. Yeah. Yeah. And you just love typing those in, didn't you? Oh, yeah. And then you get a number wrong and it's like, no, no, it's not a box. It's not a steel box. It's the plastic box. You know, sorry, I worked at Dixman Electronics, right? We sold boxes. So before I get too far ahead of myself, got to start with the basics, which is the first kind of barcode wasn't called this in the beginning, but these days it's been retrospectively relabeled as the one dimensional barcode. Of course, back in the originally it was just called a barcode. But anyway. Yeah. So in 1949, two blokes, guy called Norman Woodland and Bernard Silva filed a patent application called Classifying Apparatus and Method. And that described two different styles of barcode. One was referred to as the bullseye, the other one as the linear. And the bullseye one looked oddly like the Target logo. It was a bullseye, you know, although I'm pretty sure that Target didn't rip off this circular, the circular barcode for their logo. but still. Anyway, the linear style is, of course, the one that everyone uses and knows. Anyway, that patent was awarded in 1952, but it wasn't really until the Association of American Railroads attempted to use them on their rolling stock for tracking purposes. It was in the late 60s, early 70s, but that people sort of started to take them more seriously. But the problem was that, you know, in the real world, you've got dust, you've got wind and rain and sleet and blah, blah, blah. And inevitably, it caused read errors quite predominantly. So it was quite unreliable and they actually abandoned the project. But around about that time in the early 70s, there was a consortium, I think it was the mid 70s, the National Association of food chains, NAFC for short, in the US. And what they did was they were looking into barcodes as a method of essentially identifying grocery products, groceries, products at point of sale that you could then easily link to a price to accelerate how long, basically reduce the amount of time that it took in order for a cashier to ring up the list on the point of sale terminal of items that you're buying. And that was the main driving force because I see it as an attempt to improve efficiency because the old way it used to be done and unfortunately maybe we're both old enough to remember this was the lookup tables. Yeah. Yeah. So an item would come in and most of them were tagged, you know, the old tag system that had you would dial up the little the price, the numbers on the pricing gun and you have a reel of the tape and it had this little narrow maybe half an inch wide little sticky thing and the little sticky thing you would simply pull the trigger on the gun and it would, the little hammers would fire with the numbers that you selected and it would punch out a price and then you would in the same sort of motion you could then sort of rub it as it were against the surface of the product and that would put a price on it a price on every item. Yeah. And I miss that. Yeah, that's right. Because of course, the barcode doesn't contain the price. And so that was that was sort of the way it was done. So it would come to the checkout. And what they would have is they would have items where individual items were were labelled and priced in that fashion. And of course, the other ones that weren't, they would have lookup tables and yeah, they would look up the table and say, OK, what is it? It's it's this and it'd be microscopic, you know, this table because grocery stores are huge. They'll have thousands of different things for sale. That will depend on the size of the store, I suppose. But anyway, and I would look through there and it's, OK, yeah, right. It's a it's a bottle of tomato sauce, blah, blah, blah. Done. Or ketchup, I should say ketchup, considering the listenership. But there you go. So bottom line is it was slow and tedious. Good cashiers would know a lot of them off by heart. They'd know straight away where to look or they wouldn't even have to look. But you know, and sometimes the tags would fall off and they'd have no choice but to look it up and anyway, it's a time consuming. So anyway, the consortium said barcodes could help us. So what they did is they started to develop a standard and the standard is referred to as the UPC, which is the universal product code standard. And it applies to one dimensional barcodes or at the time anyway, it did. So that's standardized, was standardized as an 11 digit code, and that identifies any product as a unique product. It doesn't identify the product itself as, you know, if I have 10 packets of chewing gum, it doesn't actually identify them as this is chewing gum packet number 1006, this is packet number 1007. It just identifies all of them as a pack of chewing gum, probably of this flavour and of this brand. And that's pretty much it. So, obviously, you can build this into the wrapping or you can print it as a sticker. And in the early days and even still quite predominantly today, it's done as a sticker after the fact But a lot of products have it built into their actual, the actual container for whatever it is you're buying So the first one to do it was in 1974 and that was Wrigley's with their chewing gum So I don't mention chewing gum, I actually don't like chewing gum, but you know, lots of people do So there you go Anyway, so chewing gum, yes. And so Wrigley's chewing gum, they were the ones that first actually had a barcode built into the packaging. But for everyone else, of course, there were just stickers. So you would print off a bunch of stickers, stick them on and you'd be done. You'd be all good. More and more companies supported it and made it a lot easier for the people in the shops. So anyway. Okay, so the linear design, so Norman Woodland's linear design was essentially proven to be the winner for reasons that aren't immediately obvious, because the thing is that the bullseye one was actually more geometrically tolerant. And by that I mean you could project it at different angles to the scanner and it would scan more reliably from that point of view. But the downside of it, the big downside of it was that the printing technology of the time and even today tended to streak a little bit because you got to think about how printer works, right? So, you've got a head, either the head is moving or the paper is moving or both are moving. And so, at the time and even as I say, to some extent today, as it goes past, it would pick up a little bit of residual ink as it went past. So, when you had the circular ones, the bullseye style of barcodes, that would tend to cause a blur and that blur would corrupt the spacing between the circles and that would then corrupt the data. Whereas with a linear barcode, if you print the barcode on its end that is to say the direction of travel of the paper or the ink head is such that any streaking that does occur is actually along the length of the bars themselves, not across them. So you're printing it vertically? Yes, exactly. And therefore, that streaking is no longer an issue because you're streaking onto yourself, if that makes sense. That'd be a weird show title. Anyway, point is that these ones won primarily for that reason, despite the fact that they were a little bit more annoying in order to scan them, they actually worked out more reliable from end to end. Okay. So I talked a little bit about efficiency and that was the whole point of doing this rather than you know ringing them up one by one through a lookup table, people's memory and price tagging with price guns. So what were the efficiency gains I hear you all ask? Well it turns out that there was an improvement in efficiency and it doesn't sound like much but they did a lot of studies on this in the 70s and early 80s and it's about 1-2%. And you may think to yourself, well, you know, that's, that's not much, but you multiply that by every single day over multiple years. And if you're running stores and you have hundreds of stores all around the country, you know, even in one state that adds up real quick. It's a cumulative gain. Yeah, exactly. So one or 2% improvement, they'll take it. it. So that however isn't the interesting part. The interesting part is that once they did this it improved the efficiency and that had a knock-on effect. It actually increased average sales. Now if you think about it, it kind of makes sense if you're standing in line and it's taking forever for the people in front of you to get their stuff scanned through. You know, those items you got in your trolley, you're thinking, "Oh, I don't really need this." Yeah, exactly. The second guessing thing, I reckon that's where a lot of this comes from. Of course, that's what I quote unquote, "I reckon". OK, yeah, sure. So that's real scientific. But the point is that this is measurable. They measured this 10 to 12% increase in sales after the introduction of barcode scanned products. Go figure. Interesting. Very interesting. Okay. So that is in a nutshell, the motivation for doing it, a little bit of history about how we got to where we are. But of course, you know, this is pragmatic and I cannot let it pass without discussing exactly how it works. So you may think it's straightforward. Maybe it is straightforward, but it's still interesting. So there are essentially four broad types of barcode scanner, and I'm talking about one dimensional barcode scanners. The first is called the pen type or the stylus type, if you prefer. And these are the first ones that I saw in use was actually in a library and they initially were sort of common. Well, I say sort of like convenience stores, you would see them, you'd see them in like I saw them in use at... It was at university, they were using these, but in supermarket chains, they tended not to use them because a little bit more labour intensive and the sort of thing that... Anyway, you had to have a knack to use them. Okay, so the idea... Yeah, that's scary to say, there's room for operator failure. Yeah, exactly. So, okay, I'm getting ahead of myself. So what is this thing? So it's essentially it's a pen. It is a pen, usually with a wire at the back, more modern ones are wireless, and it contains a scanner and a light source built into the tip. So what you do is you drag it at a uniform speed, and that's the killer, a uniform speed across the barcode left to right, right to left doesn't really matter, I think. Actually, no, it does matter. it should be left to right across the barcode. And if you get it right, and it's tricky to get the hang of it, but if you get it that uniform speed right, then it will pick it up reliably enough. But the problem is, you know, we as you know, as human beings are really quite quite difficult with uniformity and time. So, you know, unlike a machine that will scan at a fixed rate every time and with minimal variability, the human arm and the human hand won't. It just won't. So they were prone to errors in reading, which would cause frustration because you'd read it, the number would come up and it would say, oh, you have a packet of razor blades. Oh, no, it's actually a packet of chewing gum. So the reset, scan again. Oh, you have a packet of... I was actually, no, I'm not going to say that. My point is that, you know, prone to errors. OK, let's move on before I dig myself into a hole. Laser scanner. So a rotating mirror or a set of rotating mirrors move a laser beam at a constant speed and when it comes into contact with a barcode, the light is reflected back and picked up by the sensor. Now, that idea, of course, means that the laser is scanning at a rate that is faster than an average person could move it. And this is one of the things that people, you know, when you move an item across a laser scanner, people think, "Oh, okay, you've got to scan it like a certain way. It's like I'll do a sort of like a whip in one direction. It'll be faster and I'll go this direction, that direction, whatever else." And they think, "Oh, they get the knack of scanning it." And it's like, "Oh, if I scan it in this way or at this rate or at this angle, I'll get a faster result or a better result. No, it's all about the light being reflected back and you'll hear the little beep because the beep simply means, yes, I've had a successful scan of the barcode. And if you sat the product on the bed, on the actual scanner and let the laser do its thing without moving it, you would probably get a better, more reliable result because the laser is doing the work moving it. You're moving it, it's confusing the laser. Yeah. So, you know, and if you watch the checkout operators, you know, they get very good at simply sliding it across the panel because they know that it'll pick it up as it goes across the panel. Some of the ones that you'll have problems with will be, you know, sometimes they'll have the UPC barcode printed on something with a curve on it or if it's slightly reflective or if there's a bit of dust or dirt or grit or something like that. And those sometimes you'll have condensation on cold stuff. Yes. That's a a good one. Yes, condensation is a problem and they'll have to hold it different angles to get it just so that the laser will scan it correctly. And that's, you know, it's got nothing to do with the movement. It's got to do with, you know, the laser striking it at an angle where it picks up all the whole entirety length of the barcode in one scan distortion. Take distortion. Anyway, okay, so the idea is, oh gosh, shameless. So, the idea is that, um, that idea, sorry, has been the most preferred way of doing it pretty much for the last 25 years. So, laser scan is good. Okay, CCDs. Now, there are essentially two types but the original type and still sort of popular is the single row. they're very densely packed together light sensors, they read ambient light reflected off the barcode and that's the key difference, but it's also a single strip. So the idea is it's linear in the same sense of the barcode is linear. And that's more cost-effective than the fourth type which is an imaging type, which is what most people call a camera sensor. It's a two-dimensional array of CCD elements that most people would simply call a digital camera. And it literally, it takes a photo of the barcode and then runs it through image processing software to decode it. And that is becoming the prevalent method. And it's only becoming prevalent because the decoding is becoming better, faster, cheaper, all the same sorts of reasons, lower power, all of those things. And it's also becoming more common because it's more flexible. We'll talk about why in a minute. So some of the key points, The pen and the laser essentially, they share a common problem or feature, right? They illuminate the barcode using a specific wavelength of light that's emitted by the laser. So that means that their photo detectors are specifically selected for that wavelength. And you may say, okay, that's fine, that's good, it's bad, maybe whatever. But I mean, that is still something to be aware of as a limitation of that technology. But honestly, it's the sensor that's interesting. So how it senses the light, there's an excellent link in the show notes for photosensitive devices. It's by Ray Marston and Tony Van Roon. It's really, really good if you really wanna read about how those particular photosensors, photodiodes actually work. And I encourage you to read it if you're interested. But before we dive into too much of the technical detail, I'd like to talk about our first sponsor. And that's Audible. Audible is a leading provider of premium spoken audio information and entertainment that allows listeners to choose audio versions of their favorite books. Why would you wanna do this? Well, many of our day-to-day activities, you need your eyes on the job. So when there's a book you really wanna read and you're so busy with other things, you just can't find the time, well, that's where audiobooks come in. It's much easier to multitask when you're listening to a music podcast or an audiobook. Whether you're driving, doing housework, yard work, With Audible, you can still read your favorite book and not miss out, it's really cool. You can buy books individually, or you can sign up for the Audible Listener Program, which gives you book credits each month for a low monthly fee. You can download your audio book to your PC or your Mac, your Windows phone, Android, Apple iOS device, listen to it wherever you might be. I'm a huge Douglas Adams fan, people know this, love his stuff. And when I had a look at what Audible had, There's about 20 different audio books in their library and includes all of the Hitchhiker's Guide to the Galaxy books, the Dirk Gently books, as well as the two Doctor Who episodes he was involved with, including Shader, which was the unfinished episode that never got aired in its entire world. They never finished filming it, but anyway. Anyway, it was the basis of the first Dirk Gently book, Dirk Gently's Holistic Detective Agency, with Professor Chronotis and all that. Anyway, never mind that. Read the book if you want to know more about it. Or better still, get the audiobook and listen to it. The best part though about some of these books is that they're read by different people. So some of them are actually read by Douglas Adams himself, which is, you know, unfortunately he's no longer with us and it's one of those things that it's wonderful to have that to go back to. But, you know, if you want, there's other people that read the same books as well. So you can have the same book read by a different narrator. So lots of options, lots of things to choose from. Now, Audible, they've got books in business, the classics, fiction, history, erotica and sexuality, mysteries, thrillers, sci-fi, fantasy, self-development, kids, young adults. I mean, lots and lots more to list. With over 150,000 titles, pretty much every genre that you can imagine, you're going to find what you're looking for. Now, right now, you can get a free audio book with a 30-day trial by signing up at audiblepodcast.com/pragmatic. please make sure that you use that specific URL if you wanna help the show out. Audiblepodcast.com/pragmatic for your free audio book. I'd like to personally thank Audible for sponsoring Pragmatic once again. So the photosensitive devices that I wanna talk about, nice common ones, the LDR, which stands for light dependent resistor. That's the original technology that was commonly used for the scanners. Now, sometimes they're called a photoconductive cell or a photoresistor, it depends on who you speak to. So the construction of an LDR, you have two terminals, they connect to each end of a strip of photoconductive material, and that's placed on a supporting insulating substrate, usually a ceramic of some kind. And it weaves its way from one end to the other, usually in a little snake pattern, like a zigzag of sorts. Now, different photoconductive materials respond differently to different wavelengths of light. Cadmium sulfide was the most commonly used because it responds to light wavelengths that are within the human visible spectrum and because it was relatively cheap. But I said was past tense because now cadmium is considered a heavy metal and heavy metals are the enemy and therefore it's been restricted in many, many countries now. So it's sort of fallen out of favor. But that's okay. Other technologies have come along so it's all good. But anyhow, the way that it works is that I guess, certain materials, when light, you know, photons hit them, some of that charge will free up electrons and that reduces the resistance of the material itself. So in effect, the bright light intensity, as the bright light intensity increases, the resistance decreases. So in dark environments, the resistance is higher. So these LDRs can drive a transistor amplifier or an operational amplifier aka an op amp and these can in turn then drive relays and eventually that could be useful for turning on and off things like street lamps let's say you know you want to turn on a night time and that application the other signals are very small and you're using each subsequent amplification stage to drive more and more power obviously that's not the context that I'm using mean but it's just another point that LDRs are used for that purpose or can be use for that purpose. All right, photodiodes. Now here's a little known fact, all silicon PN junctions are photosensitive. Who knew? Well, I guess obviously someone figured that out, so someone knew. Anyway, a photodiode is a conventional silicon PN junction diode, in a case, but it's got a transparent cover that allows light to reach the junction. So Real quick about silicon, you have N-doped and P-doped silicon. One has holes, one has an excess of electrons, essentially based on the different doping that's used. Now, when light hits them, then some of the charge creates holes that can carry more electrons, which then means that it reduces the resistance of the silicon at the junction itself. And when I say it creates holes, what I mean to say is that it essentially the charge transfer pushes those holes to become vacant to allow flow. It's the best way of putting it, at least. Anyway, it comes down to the fact that the energy that's imparted based on you know, the light wavelengths, the photons, that the energy imparted has to be enough to overcome the band gap for the electrons so that they can jump to the next energy level and you can actually get current flowing. And once that's achieved, it works like a charm. Now a phototransistor is the same kind of thing as a photodiode, same thing but it's a transistor so PNP or NPN and same thing it's transparent cover but the additional junction and the geometry of a transistor, how it's physically put together, that different geometry makes them far more sensitive to light than a photodiode. Okay, so CCDs and imaging. I am going to cop out right there and say I'm saving that for a future episode. Stay tuned if you care. Is there a topic all to itself? It is a topic all to itself. We'll get to it. It is on the list. Well, whether it's on the list or on the official list or on the unofficial list, I don't know, but it's on the list, believe me. So now we're going to get back just briefly to touch on. So that's roughly how the scanner, the component works. So essentially the light and the dark, you put all these things together and like essentially you can then use the individual detectors to detect if they're light or dark and that detects whether the light that's reflected off of the barcode. Hey presto, you can then digitally recreate the structure of the barcode as a representation of binary ones and zeros in the scanner which you can then convert into the actual numbers. Hey presto, you have yourself a scanner. So a few more notes about laser scanners. They are awesome because they work over longer distances, you know, up to a foot usually, quite easily. But the downside is they won't usually work with a smartphone image because, well, they're trying to reflect light off of an opaque sticker. So some of the older scanners will not work all that well. The newer CCD ones will work with anything pretty much because they're essentially taking a digital image so they'll work just fine with a smartphone image. Yeah, provided there's enough ambient light or the backlight on your smartphone is turned up high enough because there are apps that will allow you to scan in, take a photograph of your cards, whatever they might be, and you can then scan, display that and then as you swipe that over a CCD scanner, it will recognize the barcode and away you go. And that's awesome. But it doesn't work with all of the scanners so it's a little bit of a flawed in-between tech. And that leads us to the next generation, the next evolution. Two dimensional barcodes. Oh, yeah. 2D. Because like we haven't got to 3D yet, but that's OK. Maybe someday they'll do that. But for the moment, 2D is where it's at. And everyone knows the most common, well-known, popular two dimensional barcode is called, Vic? The QR code. Yes, it is. And you know what it stands for? I do not. It stands for the quick response code, which is a really dumb name, but never mind that, whatever. That's what it stands for. Let's just run with it. We accept reality and all of its imperfections because, well, we don't have many choices. OK, so there's lots of different ways of doing it, but that has turned out to be the most popular one. It was originally put together by a Toyota spin off called Denso, Denso Corporation. I think it was originally called Denso Wave back in '94. So Toyota spun them off and doing different parts of their vehicle, manufacturing, component tracking and stuff like that was what it was designed for. It was the component tracking, I should say. Either semi or fully automated vehicle assembly systems made use of the QR codes. Now it's instantly recognizable by it's a square of dots. Many of the dots will touch each other on one or more sides, but it's the three squares of which you have a solid square in the center and a square border around it. And there's three of them, one, two, three, usually on the top left, top right and bottom left sides. Although, of course, you could argue that that depends on how you want to look at the code. It's on three of the four corners of the square. And you'll know that's how you know that that is specifically as a QR code. Now, the reason that you have those, those are equidistant, they are fixed spacing. And what they do is that they allow it to allow the imager to compensate for three dimensional skewing of the image, because when you take an image, the imager cannot tell whether or not the top like the top left corner is closer to the CCD than the bottom right corner. And that causes a skewing of the data and, of course, corrupts the data. So by having squares of the identical size and identical spacing in the X and Y directions, you can compensate for any visual skewing when it's scanned. - That's cool. - Yeah, yeah, they thought about it. In fact, I just can't think about how it would work any other way, but anyway. All right, so the amount of data that you can store in a QR code symbol depends a lot on a bunch of different variables. So there's the mode, the input character set, and of course the version. The version refers to the size. So versions I think vary between one to 40, 40 being the biggest, one being the smallest. And that affects the overall dimensions of the symbol. So, you know, so for example, 40 will be, you know, can several inches, can be several inches wide, but obviously stores a heck of a lot more information, you know, which makes sense. Anyway, and there's also error correction built into it as well. and you can have differing levels of error correction based on the reliability required from the data when you scan it. The Wikipedia article is not too bad 'cause it gives you some examples. There's also another link in the show notes to the QR code website itself that talks about the standards in depth. The most common QR code that I've seen in use is the version three. So you may ask, so it stores more data. Well, how much more data compared to UPC? So we're going to take as an example, one of the most common, which is version three. That's a 29 by 29 grid type L numeric. OK, it's important to specify all of that up front because all of that affects how much data it can store. 127. And this is all based on numerical digits, not talking about alphanumeric, I'm talking about numeric digits because UPC is numeric digits. So we've got to compare apples to apples here. So again, most common barcode, one-dimensional barcode is a UPC type A. So 127 it was for the QR code. And how many for a traditional barcode? The answer is only 12. - That's significantly more. - Yeah, it's huge by comparison. So a little bit about the numbering system on a UPC. I just had, for anyone that's curious, The numbering system, the very first number you get on a one-dimensional barcode is actually refers to the numbering system. So zero is a regular, zero and seven are both regular UPC codes. A two is weighted items and they're marked at the store. So if you have a weighted item, then you put a two in front and a barcode identifies what it is. Well, let's say it's oranges and you need whatever else, that barcode will then identify it as an item that needs to be weighed, for example, and then calculate it based on that. A three, yeah, yeah, they put a lot of thought into this A three in front is the national drug or health related code just to differentiate those sorts of grocery items from the others A number four is an in-store use for non-food items and the following code structure that I'll talk about in a minute is not required, so it's kind of like a free-form flag that you can structure the remaining 11 digits however you choose well, remain 10 digits, the last digit being a check digit you can't use and a five is for coupons. The other numbers, the other preceding numbers are simply reserved for purposes that are that I am not aware of. So following that first digit, you have five digits, manufacturer code, five digits, the product code, and finally, a check digit, which as Clinton pointed out in the chat room is a checksum of sorts. Wow, it's-- yeah. So that's the structure of UPCA code. Now, QR code structures are very, very different because, of course, the data can be used to represent a whole bunch of different things because you got so many more options. The most common use of a QR code that people will come across is a URL. So you can use it alphanumerically to describe a uniform reference locator, which, you know, like techdistortion.com. But, you know, but of course, you know, link shortness, of course, being a smart way to go, you would use a link shortener. So the UPC code would describe the link shortened URL to save on alphanumeric size, and then that would take you to a promotional website and hey, presto, there you go. Pretty cool? Yep. Okay, so now we want to talk about the next evolution of identification and the next evolution of that is RFIDs. That's the trendy stuff. And before we do talk about those, however, I'd like to talk about our second sponsor for this episode and that's Mandrill. Mandrel is a scalable, reliable, and secure email infrastructure service trusted by more than 300,000 customers worldwide. That's a lot. I've been asked what is Mandrel because most people understand email newsletters showing up their inbox and a lot of them come from MailChimp, which is also an excellent service. But what's Mandrel? Well, Mandrel essentially is the foundation that MailChimp is built on and it's been broken out into its own service for discrete email transactions rather than one big mailing list like MailChimp. So you can think of them like individual transactions and hence the terminology transactional email. And that's actually how it started. In the two years leading up to 2012, Mandrill borrowed a bunch of MailChimp's best engineers and working in isolation, their little skunk works project became Mandrill. And it's now the largest email as a service platform on the market with more than 300,000 active customers, like I said before, of which I am one. Now let's say you run a website like Tech Distortion, for example, and you need to send feedback form confirmation emails, well, Mandrill can do that. See Tech Distortion uses Statomic as its CMS and I run Raven Forms as a plugin. And all you need to do is add the API key for Mandrel and that's it. It just works straight away, straight away, no other tweaking required. There's a bunch of new features on the site lately and they all like topic suggestions and so on and they all use Mandrel for all the emailers. I had been using standard PHP before that, PHP Mailer, but since I switched to Mandrel, it's been much easier to track and debug and frankly it performs better anyway, it's faster. So you can use Mandrel to send automated one-to-one emails like password resets, welcome messages, confirmations and even customized newsletters if you want to build out that sort of infrastructure on your server side. Of course, you could just use Mailchimp if you wanted to but anyway. Mandrel is quick to set up, it's easy to use and it's very stable. It's been made for developers by developers with extensive documentation, lots of different integration possibilities through their excellent API and the services are very high delivery rates as I said before, and the webhooks and Adalinks are also excellent. Now Mandrel's website has a well-organized interface and flexible template options, custom tagging and advanced tracking and reports. It's also the only email infrastructure service with a mobile app that lets you monitor delivery and troubleshoot from wherever you might be when you're out and about. Mandrel, as I said before, is really fast. It has, they have servers like that all around the world. They can deliver your email in milliseconds. Now I timed it on Tech Distortion after form submission, email shows up within a second of hitting submit. It's fast. So detailed delivery reports, advanced analytics and a friendly interface mean if you're in a large organization, the entire team from development to marketing can monitor and evaluate the email performance easily without hassling you, the developer. And that's handy, believe me. Visit mandrill.com and sign up today. You should, because it's free. No credit card, no commitment. Just sign up. Use the promo code Pragmatic when you do so to get 50,000 free email transactions per month for the first six months. That's huge. That's four times what you'd normally get. Integrate, deliver, track and analyze with email infrastructure from Mandrill. Thank you to Mandrill once again for sponsoring Pragmatic. I love Mandrill. - It's much better than the PHP mail. Oh, so much better. I love Mandrill. I mean, really, they sponsor the show. That is the ad read. But really, seriously, it's great. OK, so near field communication is technically what RFIDs utilize. However, we're not talking about PayPass, PayWave, Apple Pay, Google Wallet, NFC based technologies. and the problem with the terminology that I have is it's a bastardization, it's a conflation of terminology so NFC refers to the method of communication from a physically nearby radio interrogation to a response and it looks at it from that point of view, whereas RFID is a specific technology that's built using NFC principles So RFID uses NFC to transfer the identification information that they're looking for. So one is the technology and the other is the application of it? Yes, the other is an application that utilizes the technology. And I say the technology like NFC is a technology. Well, NFC is not a technology really. What NFC simply refers to is the fact that you have, with radio signals, signals there is what I refer to as the near field which is only a handful of wavelengths away from the source of origin of the RF energy. So near field simply means that it works in close proximity to the transmitter and it only works in close proximity to the transmitter because it's when you are in close proximity to the transmitter that you can get an effective power transfer and that'll make sense in a minute. Anyway, okay so RFID types. Let's go through the four permutations and combinations and some of them have really funny abbreviations. Let's start with Pratt. That's right, Pratt. Passive reader, active tag. Of course the next one is an active reader, passive tag. Active passive reader, active tag, passive reader, passive tag. So we'll get go through them each one at a time. So passive reader, active tag, Pratt. This system has a passive reader and it only receives the radio signals from the active tags. So in other words, your tag's got a battery in it because it's active and your reader, well, it just sits there and listens. Active reader with a passive tag. This is the by far the most common and the one we're going to focus on the most in this episode. and that is that system has an active reader that transmits an interrogation signal and it receives the authenticated reply from a passive device. How that works will make sense in a minute. OK, active reader, active tag, obviously that's a combination of the above two and passive reader, passive tag. Well, that's a failed technology since it doesn't actually do anything because it's far just it's passive aggressive. There you go. Think it through. Okay, biggest... Sorry, I just... It's the fourth combination, right? Got to mention it. No, I don't because it doesn't exist. Okay, biggest advantages of active tags is that they allow for high frequencies and longer distances. Sounds like a bonus and it can be. But the disadvantage should be obvious, they need a power source. Even if you put a long life, low drain lithium battery in these things, they're going to have a limited lifespan. And honestly, these things are mass produced. They don't want to mass produce them with a replaceable battery. So typically they don't. So they use them for two or three years, battery dies, chuck them out, get another one. That's the way that they tend to work. So anyway, and that also makes them more expensive. Now, the way a passive tag works is really kind of, I think, is kind of ingenious. So what they do is you'll know a passive RFID tag because it's about one inch square, maybe an inch and a half square. Obviously, you get larger and smaller ones, but they're about that sort of size. And you'll see a series of concentric conductive strips on this sticker, and it'll stick to the side of the product. Some people refer to them as a security tag, right? Technically, there it's an RFID tag or rather it's the same kind of technology that you can put an RFID into it. So, those conductors, the little pattern that they form, what they do is they essentially act like a coil. Now, with an electrical or magnetic system, when you've got a rotating, alternating magnetic field, electric field, then it'll induce a current in that conductor, which is exactly how a generator works. Generator spins around, it induces a current in the windings generating electricity. So same kind of idea. So the idea is that radio energy is an electromagnetic wave. So the electromagnetic wave is carrying actual energy. You can capture that through these RFID tags, through the concentric conductors that form the antenna. And that energy charges very, very small current, very, very small, not much, doesn't need to be much. It charges a very small capacitor, and that capacitor will store that energy. So what happens, it sends out a burst. That is, sorry, the interrogator, the reader, the active reader, it will send out a burst of electromagnetic radiation, referred to as a radio signal. So it'll simply say, "Hello." Well, OK, it doesn't say the word "hello", you know what I mean? Although that'd be really creepy if it did. Anyhow, so anyway, I digress. So... - What's up? Yo, RFID tags, what's happening? Hello, RFID tags, are you there? OK, that's getting more disturbing by the second. Point is, yes, that the RFID tags, then the charge in the capacitor then drives the circuit and it's a very small, very low power transmitter in the RFID in the centre of the RFID sticker, and that will then respond. So it then transmits its response and its response is of course its own identification So it'll respond with a radio signal and modulated essentially with its data which is its own identification - Who goes there? - Yeah, exactly And it responds with its name, who it is, what it is So the interrogation signals is on for a brief period and then the response signal is on for a slightly longer period but it's very low power response. And because the device is only able to be powered by whatever charge it accumulated from the radio signal from the active reader, from the interrogation signal. Okay, so far so good? Okay, cool. Obviously, RFIDs are used in a whole bunch of different applications. Now, the one we're talking about specifically is identification of goods at the point of sale. But there's lots more. Ones we aren't going to specifically talk about, but it's still interesting and worth mentioning is tracking of goods in a warehouse, access management, that is to say access control. And I have RFID tags on my security badge. So my security badge, when I go into my building at my place of employment and my job, I swipe the card and it has an RFID in it and it responds with, "Yep, it's John, you should let him in." Or, you know, if it's the kitchen area, don't let this guy in, he looks like he's gonna steal all the rest of the mints. Anyway, access management. So there you go. Tracking of persons or animals. So you, sometimes they'll put RFID chips in a very small injectable tube that can be injected underneath the skin and you can literally track, like, you know, when they chip a dog and that or a cat or whatever, you know, that is an RFID tag and that is injected under the skin and, you know, that allows you to be tracked, which is kind of, you know, raises all sorts of other interesting issues we're not going to go into this episode, but never mind that. Toll collection. So you drive in your car, you've got a transponder in your car, you can drive through the electricity. Yeah, that sort of thing. Speed pass. Here in Brisbane, it's referred to as, well, not just Brisbane, but in different parts of Australia, you've got eToll, ePass and Goviya. There's a whole bunch of different names of different consortiums, and they're all subtly different, but they're all the same technology, essentially. You know, contactless payments like your Visa card or your MasterCard or more, you know, your new shiny new iPhone 6 or 6 Plus. Apple Pay. Apple Pay and your Apple Watch that hasn't been released yet. or of course if you've had an Android phone for years, you know, in them. So, well, not all of them, but some of them. So contactless payments, travel documents, you know, airport baggage tracking. So there's so many places where RFIDs are used. But we're only really interested in point of sale for the purposes of this discussion, I think it's best to focus on the one component. Okay, so the major advantages are unlike a barcode, you could scan multiple RFID tags at the same time. That's a time saver. That's a win. Oh, that's a big win. The other major advantage, RFIDs can be completely covered as in physically covered and they don't have to be visible to the scanner and you can still scan them. That's a big advantage, too. More data, too. Oh, yes, sure. That's true. Yes, that is another advantage quite right. Major disadvantages. In 2011, the average cost of a mass produced passive tag was nine cents US each. Now, if I do custom UPC labels, because I couldn't get the and I couldn't find an equivalent mass produced cost, but if you do limited production runs of a thousand or more UPC labels. These are standard UPC one-dimensional barcodes. You're looking at three cents each US. Now that is, you know, obviously one third of the price. Now, obviously it's also printed many times it's printed onto the product at the time of manufacture. but there are some stores where they want to override that for their own reasons or you know, there's a whole bunch of reasons why they may want to. Irrespective, the point is that it is significantly cheaper. It doesn't sound like a lot of money, but honestly, if you're not getting much... - Cumulative. - Yeah, it accumulates, it's a lot. And if you're not getting much of an efficiency gain from the expense, then why are you doing it? The disadvantage anyway, I'll get more about some of the other advantages in a minute, but the other disadvantages are that it's more fragile than a barcode. So if you get it wet, you can kill it. Static discharge will kill them. Nearby lightning strikes and power surges can kill them because they're designed to accept and amplify a radio wave and capture all of that energy. Well, it's really good at capturing static discharge. And of course the EM pulse from a lightning strike as well. And that'll have such a high voltage because it's uncontrolled that it can fry the RFID chip and blow the cap. So, you know, not good. None of those issues affect barcodes. Okay. So all of that's all well and good, I hear you say, but let's talk about the whole idea of the whole multiple things at once. Where could we get to? What's the real big benefit from using RFIDs in a grocery scenario? Well, before we get into that, think about the cost problem, which is the passive tags that I quoted, the whole nine cents each. Those are the cheapest passive tags that you can get. Now, they only have a small sensing range. Now, the newer technology, the newer technology, well, that's, you know, that can scan multiple metres, you know, 10 feet away, let's say. But the cheap ones that I just mentioned, the ones that are potentially borderline cost effective They have a very short sensing range Now, you can increase the size of the tag to improve that But obviously that increases the size of the antenna and that increases the cost And before you know it, it's just all over Red Rover, right? Why are you doing it? It's just not worth it anymore So, and the problem is simple, right? As I discussed previously on both episodes 16 and 39, you're up against the inverse square law RF travels in an ever-expanding bubble and that sphere, the energy decreases rapidly as that expands in terms of the distance away from the source of the RF energy from the actual active reader So it should be obvious, you have to go to a bigger one to get further away Again I refer you back to the advantages of the active readers with the actual battery power source in them them. Of course, you can go further away because you don't need to project as much energy. And remember, you don't want to be projecting an infinite amount of energy because if you're projecting an infinite amount of energy, you're going to fry the poor person standing next to it. But it's true. You know, those PayPal, those eToll things, don't stand in front of them for eight hours in a day, like at the side of the motorway. Don't do that. Bad idea. Anyway, all right, it's like standing next to a mobile phone tower at like 10 feet away. It's also dumb. Don't do that either. Okay. So as costs come down, ultimately, where we want to go with this technology is, here's the concept, okay? You put everything you want in your shopping trolley. You walk through the exit. RFID scanners read everything in the trolley. it automatically charges your paywave, paypass, Apple pay, payment method of choice wirelessly and you just keep on walking out the door, done. Done and dusted. That's the pipe dream. Well, that's the dream. Whether it turns out to be a pipe dream, it remains to be seen. But, you know, and it's not a reality yet, but it's getting closer. So, as a slight tangent, just briefly, When I lived in Calgary, and this is over a decade ago now, the Calgary Co-op was trialing a scan as you shop experiment. I don't know if they're still doing it, but I know that Tesco's are doing it. And the idea is that as you put an item in the shopping trolley, you scan it with a portable battery powered scanner as you go. So as you add an item, you scan it, put it in the trolley, scan it, put it in the trolley. It keeps a tally of how many of each quantity, so you scan each wine. If you got 10 individually wrapped biscuits, well, sorry, cookies, whatever, you would scan all 10 of them. So it keeps tally of exactly what's in the trolley. And it's all linked into their pricing system and can give you a running tally. It's fantastic. However, the downside is it's all about trust and honesty. And we're all trustworthy and honest, aren't we, Vic? Oh, yeah, everybody. I'm like totally 100% honest. Anyhow, except for that statement. So, you know, that's what the way they got around this at the at the co-op was that they would do a random check. So you would you would log on to the system, you would swipe your card, and it would say, yes, oh, hey, you're John Chigi. And yes, we you've been here three times before. And it kept a tally of how many times you'd use the system autonomously and how many times you had not had used the system without having essentially a shopping trolley check at the end. What they would do is they would flag you the first couple of times to see if there are any discrepancies. Even though you scanned it as it went in the trolley, they would direct you over to another aisle and they would say, "Yeah, we'll just manually put you through anyway," and they would do a tally and you would build up a history of trust. It was also based on dollar value. If you only had $50 worth on your tally list and they'd had a look in the trolley and they that looks about $50 worth based on the cursory glance, then they would tend not to check you. However, obviously, there's ways of gaming that system. I mean, you can hide more expensive items inside and amongst the larger, cheaper items to make it appear as though you're good. They just say, "Oh, I'm sorry if they do pick you out. I forgot to scan those LED light bulbs that were hiding in the cereal box for some reason. But anyhow, you know, it's but you can game the system, right? Yeah. So it's not ideal. But RFIDs would use exactly the same idea, but it removes that trust issue because you can scan the whole trolley and determine what's in it. They're going to find your light bulbs in the cereal. They're going to find the damn light bulbs. So this is the problem. Although there's ways of gaming that system too, you could remove the RFID tag from the packaging. But you can't with a Faraday cage. Well, that's actually interesting because that's one of the points. If you have a metal trolley, that's going to interfere with the RFID signals because it is a Faraday cage, at least on, you know, five out of the six sides, unless your shopping trolley, your metal shopping trolley also has a lid, which I haven't come across one yet, but maybe there is one. I don't think so. So, the plastic trolleys, which I have seen in some stores and they're still they're relatively common enough, they would be the choice if you're using RFIDs, you'd go with plastic trolleys. You've used plastic trolleys before, I take it? Yeah, the metal is more common, though. Yes, you're right. Metal is far more common because it's far more robust. You are talking about what we call a shopping cart here, right? Sorry. Yes. Sorry. Shopping cart, shopping trolley. Yes. Same thing. Different name. but yes, shopping cart, indeed. Okay, but of course, all of that presupposes that the RFID signals that you're reading are exhaustive and correct. And up until now, they haven't been. They only recently got up to around about the 99% accuracy. And you got to realize if you got a bunch of RFID tags in physically proximate to each other, like literally only millimeters apart, that causes issues because they're only like - Signal noise issues. - Yeah, interference issues. So there's all sorts of ways that they have to handle it. Cause I say, oh yeah, it's possible to read them all at once. Well, yeah, sort of it is, you know, kinda. What you want to do is you want to avoid multiple people talking over the top of each other. I'm not saying multiple people, I'm talking about people. The RFIDs are people, they're little people. They're not actually little people, they're little RFIDs. The point is they're all talking over each other at the same time and that's a bad thing 'cause you can't separate them out and then everyone's talking at once and you're like, well, who said what? Who are you? What are you? I don't know anymore. So you just give up and it leads to an approach where you would stagger. So you would say, okay, I'm going to scan all of these fellows. So only people that start with these numbers respond, these numbers respond and so on. And you'll cycle through in a group based approach and that'll reduce the statistical probability of having a hundred responses at once. Now there's multiple ways of overcoming that depending upon the radio technology that they use and so on and so forth. I'm not gonna get into all of the details, but there's ways of overcoming it, but they are more expensive. They are becoming cheaper. It's certainly not impossible and it is improving all the time. And I guess that's the message. As RFID technology improves further, it becomes more sensitive, it becomes cheaper and more reliable as to say there are less missed items in each trolley, then I can actually see this happening. In fact, I'd wager there's probably still some in the world that are already trialing this. And if there isn't, it's probably only five or 10 years away from becoming a commonly trialed or even a common thing for certain supermarket chains. Because think about it, you free up checkout attendance, it's worth the investment. Employees cost money. They cost a lot of money, especially when they get sick. And you know, how dare you get sick? I just, you know, honestly, from an employer point of view, forget that people have any rights whatsoever to live a life, right? Employees suck, right? They're not productive all the time. They stand around and they talk to other employees. They get sick, they talk, they make mistakes, you know, they cost you money. Some of them are, you know, dishonest and take money out of the tills, you know, I mean, there's all sorts of issues with employees. So if you have RFIDs, a lot of those problems just disappear. You don't need the checkout attendant in order to scan the items anymore, you just push the trolley through, you need to move some of the attendants back in the chain in the store. So you need to have someone manning the deli and the fresh food area, you know, or the bakery or whatever parts of the store if your shop does have those areas, of course, some don't. But, you know, because then they would have to put them in a bag, program a label, put the label on the bag and say, "There you go, sir." You know, so that so the RFID would count. But that could save grocery shops, a supermarkets a lot in wages, and that alone makes it worth exploring. And honestly, that's all I had. Did you have anything else you wanted to add to the discussion at all, Vic? No, that covers most of it. Well, there you go. So thank you again very much to Florian for that suggestion. I think that's very, there's a very interesting topic. I hope I've answered that. Good topic. Gared is a good topic and I would strongly encourage listeners that want to contribute to the show and have things that they want me to talk about, please suggest the topics. And if I agree, I think it's a good topic, I'll put it out there and people will vote on it and we'll pick from the list. And that's how this one was added to the list. So thank you again. If you want to talk more about this, you can reach me on Twitter @johncidji. And you can see my writing and this podcast and others I've made are hosted at my site If you'd like to get in touch with Vic, how can they reach you, Vic? They can find me on Twitter @vichudson1. There will be a new episode of AppStory Podcast tomorrow for Thanksgiving here in the United States, also known as Turkey Day. Fantastic. Cool. Can't wait. Yeah. I'm looking forward to getting it back on. It's been on pause for a while. Is it a surprise who the guest is? Oh, it's Simon Wolf and we're talking about his app, AvTag, which should be of some interest to some people. There's a little buzz on Twitter, you know, about people wanting to find ways to tag and add metadata to shows and share pieces of them and stuff like that. It's an app that does that, so. No, fantastic, excellent. Well, there'll be a link in the show notes to that episode as well, so check it out. Yeah. So if you'd like to send any feedback, please use the feedback form on the website. That's where you'll also find the show notes the episode under Podcasts Pragmatic and as for the topic for this episode, if you would like me to cover any topic at all, you can suggest it and then other people will vote on it at the site under Topics once you sign up for a free account at the site. I've also started to release excerpts from the show that are off topic and a cut from the main episode and I'm calling it Addenda. Look for it on the site under Podcasts Addenda. You can also follow Pragmatic Show on Twitter to see show announcements and other related stuff. final thank you to our two sponsors for this episode and they are Audible. Please make sure you visit this URL, audiblepodcast.com/pragmatic for your free audiobook. Go check them out today. And also thank you to Mandrel for sponsoring the show. If you're looking to improve your site or app and need transactional email that's reliable, integrates easily and provides easy tracking and analysis, then Mandrel can help. Visit mandrel.com and sign up today. Use the promo code pragmatic to get 50,000 free email transactions per month for the first six months. Integrate, deliver, track, and analyze with email infrastructure from Mandrill. Thank you again to them. Thank you to you for listening. And of course, thank you to Vic. You're welcome. [MUSIC PLAYING] (upbeat music) (upbeat music) (upbeat music) [MUSIC PLAYING] (upbeat music) (upbeat music) (upbeat music) (upbeat music) (upbeat music) (upbeat music) [MUSIC PLAYING] [MUSIC PLAYING] (upbeat music) [Music] (explosion)