1 00:00:03,360 --> 00:00:06,440 Speaker 1: Get in touch with technology with tech stuff from half 2 00:00:06,480 --> 00:00:14,200 Speaker 1: stuff dot com either everyone, and welcome to text stuff. 3 00:00:14,240 --> 00:00:17,479 Speaker 1: I'm Jonathan Strickland and I'm Lauren. Before we get into this, 4 00:00:17,560 --> 00:00:21,119 Speaker 1: I just want to alert listeners ahead of time. I've 5 00:00:21,160 --> 00:00:24,440 Speaker 1: got a cold, so I'm going to sound a little weird. 6 00:00:24,840 --> 00:00:28,360 Speaker 1: He's kind of raspy. Is it's a little bit tom? Wait? Yeah, 7 00:00:28,400 --> 00:00:31,240 Speaker 1: you know the piano has been drinking and Jonathan has 8 00:00:31,280 --> 00:00:34,560 Speaker 1: been on an airplane. So that's where we're at. Today's 9 00:00:34,640 --> 00:00:38,680 Speaker 1: episode is brought to us courtesy of a little listener mail. 10 00:00:39,280 --> 00:00:43,120 Speaker 1: This one is from Daniel via actual email. Yeah, apparently 11 00:00:43,280 --> 00:00:45,880 Speaker 1: people still use that. By the way, guys, we've seen 12 00:00:46,000 --> 00:00:49,960 Speaker 1: a big upswing and listener and email and Facebook and Twitter. 13 00:00:50,400 --> 00:00:52,880 Speaker 1: We love it. Keep up the great so so much. 14 00:00:52,920 --> 00:00:55,400 Speaker 1: We're sending we We love every single note that we receive. 15 00:00:55,480 --> 00:00:58,240 Speaker 1: Even if we are too incredibly stressed out to respond 16 00:00:58,280 --> 00:01:00,000 Speaker 1: to all of you. We love you, Thank you so much. 17 00:01:00,040 --> 00:01:02,360 Speaker 1: I'm trying desperately to get through all the backlogs, so 18 00:01:02,480 --> 00:01:04,759 Speaker 1: if you have written, I am working my way back 19 00:01:04,800 --> 00:01:08,560 Speaker 1: to you, babe. Anyway, Daniel, he wrote to us to say, 20 00:01:08,760 --> 00:01:10,360 Speaker 1: I don't know if you did this before, but we've 21 00:01:10,400 --> 00:01:13,600 Speaker 1: recently had the news then online gaming streaming service called 22 00:01:13,680 --> 00:01:17,360 Speaker 1: Twitch is surpassing live streaming services like you Stream, w 23 00:01:17,560 --> 00:01:22,080 Speaker 1: w E Network live Stream, and other streaming services by volume. However, 24 00:01:22,160 --> 00:01:24,640 Speaker 1: I don't really hear about it outside of gamer groups, 25 00:01:24,680 --> 00:01:27,039 Speaker 1: which is odd to me. Can you all get into 26 00:01:27,080 --> 00:01:30,000 Speaker 1: the details about how Twitch works, or maybe into live 27 00:01:30,080 --> 00:01:33,280 Speaker 1: streaming in general. It's a growing trend in recent years. 28 00:01:33,360 --> 00:01:36,360 Speaker 1: I can only see getting bigger as gaming audiences grow. 29 00:01:36,840 --> 00:01:39,360 Speaker 1: You know what, Daniel, We agree it's a great topic. 30 00:01:39,440 --> 00:01:42,560 Speaker 1: It's one that's become incredibly important as of the week. 31 00:01:42,560 --> 00:01:45,840 Speaker 1: We're recording this podcast because they're Okay. So there's this 32 00:01:45,920 --> 00:01:47,960 Speaker 1: rumor going around, and at this point it's a very 33 00:01:48,000 --> 00:01:52,160 Speaker 1: strong rumor, although the parties that are involved are still 34 00:01:52,240 --> 00:01:55,240 Speaker 1: not commenting officially on it, that Google in the form 35 00:01:55,280 --> 00:02:00,320 Speaker 1: of YouTube is interested in purchasing Twitch for a billion dollars. 36 00:02:00,360 --> 00:02:03,880 Speaker 1: So this was a story broken by Variety originally, and 37 00:02:03,920 --> 00:02:07,360 Speaker 1: then it was corroborated by the Wall Street Journal, although 38 00:02:07,360 --> 00:02:11,520 Speaker 1: again no one from either Twitch nor Google is is 39 00:02:11,600 --> 00:02:14,639 Speaker 1: commenting on the subject. By the time this podcast comes out, 40 00:02:14,880 --> 00:02:16,840 Speaker 1: it may very well be that all of this is 41 00:02:17,240 --> 00:02:20,320 Speaker 1: supremely old news and that the the deal has either 42 00:02:20,440 --> 00:02:24,720 Speaker 1: been agreed upon or declined, but we don't as we're 43 00:02:24,760 --> 00:02:27,480 Speaker 1: recording this, don't know the outcome al right where we 44 00:02:27,560 --> 00:02:32,160 Speaker 1: are recording this on May twenty and um, and yeah, 45 00:02:32,200 --> 00:02:34,040 Speaker 1: these these two sources. By the way, Variety in the 46 00:02:34,040 --> 00:02:37,480 Speaker 1: Wall Street Journal reported different stories, um that the sale 47 00:02:37,639 --> 00:02:40,480 Speaker 1: is already certain, or that the talks are in early stages, 48 00:02:40,560 --> 00:02:43,160 Speaker 1: and that Twitch might just sell stake in their company 49 00:02:43,240 --> 00:02:45,560 Speaker 1: instead of selling the entire company as a whole. Right, 50 00:02:45,760 --> 00:02:49,280 Speaker 1: And even though official sources are not doing any talking 51 00:02:49,320 --> 00:02:51,600 Speaker 1: about it right now, anonymous sources have done a lot 52 00:02:51,639 --> 00:02:54,880 Speaker 1: of talking. For for example, Verage reported that other companies, 53 00:02:54,880 --> 00:02:59,280 Speaker 1: including Microsoft we're also according Twitch, but have been turned down. Um. 54 00:02:59,320 --> 00:03:01,960 Speaker 1: At any rate, It's it's a really interesting news item 55 00:03:01,960 --> 00:03:05,720 Speaker 1: for for both the gaming industry and the live streaming industry. 56 00:03:05,800 --> 00:03:08,640 Speaker 1: And I guess video on the internet as a whole, yeah, 57 00:03:08,800 --> 00:03:11,000 Speaker 1: And keeping in mind that video on the internet is 58 00:03:11,000 --> 00:03:14,240 Speaker 1: still a very young thing. I mean, it wasn't that 59 00:03:14,320 --> 00:03:16,560 Speaker 1: long ago that you didn't have any sort of video 60 00:03:16,600 --> 00:03:19,080 Speaker 1: hosting services. And the way you would get video on 61 00:03:19,120 --> 00:03:21,960 Speaker 1: the internet is that someone would save a video file 62 00:03:22,280 --> 00:03:24,760 Speaker 1: and you'd have to download the whole thing to watch it. 63 00:03:25,160 --> 00:03:27,680 Speaker 1: So this is a this is kind of showing the 64 00:03:27,720 --> 00:03:32,280 Speaker 1: next step in in the the world of video online. 65 00:03:32,320 --> 00:03:35,560 Speaker 1: So we've got the prerecorded stuff, which YouTube has got 66 00:03:35,640 --> 00:03:38,560 Speaker 1: pretty sewn up at least for user generated. Then you've 67 00:03:38,560 --> 00:03:41,440 Speaker 1: got the pre record stuff that is made by studios 68 00:03:41,440 --> 00:03:44,200 Speaker 1: and television networks. Now that's all over the place like Netflix, Hulu, 69 00:03:44,240 --> 00:03:47,240 Speaker 1: all that kind of stuff. And then you have broadcast 70 00:03:47,280 --> 00:03:50,080 Speaker 1: where you're doing a live streaming event, which is what 71 00:03:50,160 --> 00:03:53,400 Speaker 1: Twitch and other services like Twitch are all about. So 72 00:03:53,800 --> 00:03:56,119 Speaker 1: how does that work? Why is it such a big 73 00:03:56,120 --> 00:03:59,200 Speaker 1: deal of why why would Google even be interested in it? 74 00:03:59,320 --> 00:04:01,800 Speaker 1: A lettle in one billion dollars worth of interest in Yeah, 75 00:04:01,880 --> 00:04:03,640 Speaker 1: So we thought we would first go into a little 76 00:04:03,640 --> 00:04:06,240 Speaker 1: bit of history. We like to always look back into 77 00:04:06,280 --> 00:04:09,920 Speaker 1: how things became what they are. So let's, uh, let's 78 00:04:09,960 --> 00:04:11,880 Speaker 1: just think we're not going to get into the way 79 00:04:11,880 --> 00:04:14,880 Speaker 1: back machine because this is pretty recent history. Lauren and 80 00:04:14,920 --> 00:04:17,839 Speaker 1: I both lived through this history. It wasn't going on 81 00:04:17,880 --> 00:04:20,760 Speaker 1: immediately around us, but we were technically alive while it 82 00:04:20,839 --> 00:04:22,920 Speaker 1: was happening, yes, which is relatively rare for the two 83 00:04:23,000 --> 00:04:24,440 Speaker 1: of us. For lots of the history that we do 84 00:04:24,480 --> 00:04:26,800 Speaker 1: on this show. Yeah, there was a period in when 85 00:04:26,839 --> 00:04:29,680 Speaker 1: I was dead to the world. So anyway, Twitch is 86 00:04:29,839 --> 00:04:31,560 Speaker 1: kind of a first of all, it's a spinoff of 87 00:04:31,560 --> 00:04:35,400 Speaker 1: another company. It's so in order to understand where Twitch 88 00:04:35,440 --> 00:04:37,360 Speaker 1: came from, you have to go back a little further 89 00:04:37,400 --> 00:04:40,080 Speaker 1: and look at the origin of this other company, Justin 90 00:04:40,160 --> 00:04:44,520 Speaker 1: TV exactly who, which was founded by Justin Cohn, Emmett Sheer, 91 00:04:44,640 --> 00:04:48,479 Speaker 1: Michael Cybel, and Kyle Vauched. And so here's the thing. 92 00:04:48,920 --> 00:04:53,120 Speaker 1: Even that company was the product of some other some 93 00:04:53,160 --> 00:04:55,960 Speaker 1: other work done by this group of people that didn't 94 00:04:55,960 --> 00:04:58,320 Speaker 1: work out so much. You see, you had this group 95 00:04:58,360 --> 00:05:03,479 Speaker 1: who created a web calendar software called Keiko. This was 96 00:05:03,560 --> 00:05:06,080 Speaker 1: from a group who had gone to Yale together I think, 97 00:05:06,160 --> 00:05:10,760 Speaker 1: and formed the company soon after graduation. And so they 98 00:05:11,080 --> 00:05:16,320 Speaker 1: were approached by a brand new seed company called y Combinator. 99 00:05:16,320 --> 00:05:20,000 Speaker 1: And a seed company is all about finding startups to 100 00:05:20,000 --> 00:05:24,400 Speaker 1: to finance, to fund, hoping that those startups will blossom 101 00:05:24,560 --> 00:05:28,080 Speaker 1: and therefore those investments will end up making huge returns 102 00:05:28,160 --> 00:05:31,520 Speaker 1: in return. You also see these companies that otherwise might 103 00:05:31,560 --> 00:05:35,520 Speaker 1: not have a chance grow into something really interesting. Right. 104 00:05:35,560 --> 00:05:39,040 Speaker 1: Part of that same batch was both Reddit and Infogami, 105 00:05:39,120 --> 00:05:40,880 Speaker 1: which is one of those companies that was started by 106 00:05:40,880 --> 00:05:44,160 Speaker 1: Aaron Schwartz. Yeah, so amazing group of people, right, So 107 00:05:44,240 --> 00:05:47,680 Speaker 1: not just the smarty pants folks behind Kiko, but these 108 00:05:47,720 --> 00:05:50,320 Speaker 1: other groups as well. I mean, Reddit is now one 109 00:05:50,360 --> 00:05:53,840 Speaker 1: of the more popular sites on the Internet has a 110 00:05:53,880 --> 00:05:57,320 Speaker 1: flourishing community, although it's a community that has its own 111 00:05:57,440 --> 00:06:00,599 Speaker 1: uh issues and in fighting. Well yeah, I think the 112 00:06:00,640 --> 00:06:04,240 Speaker 1: more popular Reddit gets, the more irritated the hardcore audience. 113 00:06:05,279 --> 00:06:07,880 Speaker 1: But the same thing happened to dig so it'll be 114 00:06:07,880 --> 00:06:09,560 Speaker 1: interesting to see what happens with Reddit. But that's a 115 00:06:09,600 --> 00:06:14,160 Speaker 1: totally different podcast. So here's the thing about kiko um. 116 00:06:14,200 --> 00:06:18,120 Speaker 1: It was, according to Justin con himself, not that great 117 00:06:18,160 --> 00:06:20,560 Speaker 1: of a product, and the reason he gave behind that 118 00:06:20,640 --> 00:06:24,320 Speaker 1: was that he and his fellow programmers didn't really need 119 00:06:24,360 --> 00:06:26,880 Speaker 1: a calendar. He said, I was going to college, I 120 00:06:26,920 --> 00:06:30,360 Speaker 1: had two classes, I had classes on on two days 121 00:06:30,400 --> 00:06:33,200 Speaker 1: of the week, and otherwise I was sitting around programming. 122 00:06:33,279 --> 00:06:35,680 Speaker 1: So what the heck do I need a schedule for? 123 00:06:36,240 --> 00:06:38,520 Speaker 1: It does not useful to me. So since I didn't 124 00:06:38,560 --> 00:06:40,880 Speaker 1: need it, I wasn't using it, and since I don't 125 00:06:41,080 --> 00:06:45,520 Speaker 1: use it, the products we were making weren't very compelling. 126 00:06:46,360 --> 00:06:49,040 Speaker 1: And then Google calendar comes out and everyone's like, well, 127 00:06:49,880 --> 00:06:51,640 Speaker 1: so they decided what we're gonna do is we're gonna 128 00:06:51,680 --> 00:06:53,840 Speaker 1: sell this company for as much as we can so 129 00:06:53,880 --> 00:06:56,640 Speaker 1: that we can pay back our investors and then constrate 130 00:06:56,800 --> 00:06:59,160 Speaker 1: something else. So of course, they sold their company the 131 00:06:59,200 --> 00:07:03,680 Speaker 1: way any sane person would. They listed it on eBay, 132 00:07:03,920 --> 00:07:06,719 Speaker 1: and it's sold on eBay for some two and fifty 133 00:07:06,720 --> 00:07:09,800 Speaker 1: eight thousand dollars to a Canadian company called two Cows 134 00:07:10,320 --> 00:07:13,280 Speaker 1: to Cows. Two Cows. Yeah, yeah, they have two cows, 135 00:07:13,320 --> 00:07:15,360 Speaker 1: twice as many cows as someone who has one cow. 136 00:07:15,960 --> 00:07:19,920 Speaker 1: So yeah, except its spelled t u c ows, but yeah, 137 00:07:20,000 --> 00:07:22,160 Speaker 1: it's it's It was funny because they were going for 138 00:07:22,200 --> 00:07:25,720 Speaker 1: just fifty dollars. They wanted fifty grand. They were hoping 139 00:07:25,880 --> 00:07:28,120 Speaker 1: they could get fifty grand so they could pay off 140 00:07:28,160 --> 00:07:30,800 Speaker 1: their investors, and then they saw that it had hit 141 00:07:30,840 --> 00:07:33,000 Speaker 1: eighty grand and they thought, oh, we got some walk 142 00:07:33,040 --> 00:07:35,800 Speaker 1: in money now, and it finally sold for two eight 143 00:07:35,840 --> 00:07:39,520 Speaker 1: thousand dollars on eBay. All right, yeah, yeah, two two 144 00:07:39,520 --> 00:07:42,400 Speaker 1: cows and a calendar app that. Yeah, that's great. I 145 00:07:42,440 --> 00:07:45,000 Speaker 1: had no idea you could actually list a company on eBay, 146 00:07:45,040 --> 00:07:48,360 Speaker 1: and the idea that that's involved in the terms of 147 00:07:48,400 --> 00:07:52,520 Speaker 1: service blows my mind. Anyway, they successfully sell Keiko, and 148 00:07:52,520 --> 00:07:56,840 Speaker 1: in two thousand seven, Justin and Emmett pitched this idea 149 00:07:56,920 --> 00:08:00,200 Speaker 1: of justin dot TV to Paul Graham. Paul Graham was 150 00:08:00,240 --> 00:08:04,920 Speaker 1: the guy who founded the y Combinator Seed company, and 151 00:08:05,040 --> 00:08:07,920 Speaker 1: the premise originally was not for some sort of live 152 00:08:07,960 --> 00:08:11,840 Speaker 1: broadcast uh service for the general public, all right, It 153 00:08:11,880 --> 00:08:13,720 Speaker 1: was sort of a curiosity. The idea was to just 154 00:08:13,840 --> 00:08:19,040 Speaker 1: put a webcam um basically on on Kan's head and 155 00:08:19,560 --> 00:08:22,120 Speaker 1: have it broadcast twenty four hours a day, seven days 156 00:08:22,120 --> 00:08:23,760 Speaker 1: a week. So I think that he was doing so 157 00:08:23,800 --> 00:08:26,680 Speaker 1: it was justin TV, justin TV, right, it was. It 158 00:08:26,760 --> 00:08:28,720 Speaker 1: was like it was a kind of life casting, except 159 00:08:28,720 --> 00:08:31,120 Speaker 1: instead of having the camera turned back on himself, it 160 00:08:31,200 --> 00:08:35,240 Speaker 1: was more or less from his perspective and view. Yeah, yeah, 161 00:08:35,280 --> 00:08:38,800 Speaker 1: there's some great stories about weird things that happening. Within 162 00:08:38,880 --> 00:08:42,600 Speaker 1: the first week of going live, apparently his audience felt 163 00:08:42,600 --> 00:08:46,960 Speaker 1: that it was not the most compelling of of entertainment, 164 00:08:47,280 --> 00:08:50,480 Speaker 1: that watching a guy sit in an apartment and code 165 00:08:50,920 --> 00:08:54,040 Speaker 1: was not necessarily the most interesting subject matter, that it 166 00:08:54,080 --> 00:08:56,280 Speaker 1: needed a little bit of spice, so um, so they 167 00:08:56,360 --> 00:08:58,480 Speaker 1: provided it to the audience. Provided the spice this is 168 00:08:58,520 --> 00:08:59,680 Speaker 1: where you know, it's kind of like when you hear 169 00:08:59,720 --> 00:09:03,680 Speaker 1: about a news outlet making the news. In this case, 170 00:09:03,720 --> 00:09:06,520 Speaker 1: it was the audience making the contact content. How do 171 00:09:06,600 --> 00:09:10,319 Speaker 1: they make the content? They contact the police and say 172 00:09:10,360 --> 00:09:13,640 Speaker 1: that there was a report of a stabbing incident in 173 00:09:14,160 --> 00:09:17,240 Speaker 1: Justin Kahn's apartment. So the police pay a little visit. 174 00:09:17,280 --> 00:09:20,679 Speaker 1: By the way, this entire incident is preserved on YouTube 175 00:09:20,720 --> 00:09:22,880 Speaker 1: if you want to watch it. And so you hear 176 00:09:23,480 --> 00:09:26,960 Speaker 1: the cops knocking at the door, they come inside. They 177 00:09:27,040 --> 00:09:30,440 Speaker 1: explained that they got a report about stabbing. Pretty quickly, 178 00:09:30,480 --> 00:09:34,880 Speaker 1: Justin figures out, Hey, the it's our users who have 179 00:09:35,000 --> 00:09:37,680 Speaker 1: decided that this is this would be entertaining to watch. 180 00:09:38,120 --> 00:09:42,440 Speaker 1: What a bunch of scamps um. There were other incidents 181 00:09:42,440 --> 00:09:45,120 Speaker 1: that were similar, where they would call in fake fires 182 00:09:45,360 --> 00:09:47,680 Speaker 1: and there were like six or seven fire trucks lined 183 00:09:47,720 --> 00:09:52,080 Speaker 1: outside of Justin's apartment legally culpable scamps that. Yeah, So 184 00:09:52,200 --> 00:09:56,200 Speaker 1: eventually they decided, hey, you know this, Uh, this idea 185 00:09:56,840 --> 00:10:00,440 Speaker 1: isn't working because Justin's just not interesting enough off to 186 00:10:00,520 --> 00:10:03,439 Speaker 1: keep everyone's attention. Uh. There are other incidents as well. 187 00:10:03,480 --> 00:10:06,840 Speaker 1: There are plenty of videos online that explained the hijinks 188 00:10:06,880 --> 00:10:09,959 Speaker 1: these guys got up to But what happened was they 189 00:10:10,000 --> 00:10:13,959 Speaker 1: decided this isn't sustainable, it's not entertaining, So what can 190 00:10:14,040 --> 00:10:17,640 Speaker 1: we do. How can we modify what we've built into 191 00:10:17,720 --> 00:10:22,160 Speaker 1: something that actually resonates with people. So they revisit the 192 00:10:22,200 --> 00:10:25,240 Speaker 1: idea and they discussed different approaches, and they came up 193 00:10:25,280 --> 00:10:28,600 Speaker 1: with this idea of using the justin TV as a 194 00:10:28,640 --> 00:10:33,200 Speaker 1: platform for other people to broadcast things live over the Internet. 195 00:10:33,720 --> 00:10:37,760 Speaker 1: So this was the democratization of live broadcast. YouTube had 196 00:10:37,800 --> 00:10:43,280 Speaker 1: created the democratization of of sharing video. But but you didn't. 197 00:10:43,320 --> 00:10:45,560 Speaker 1: You couldn't do live events on YouTube at the time, 198 00:10:46,240 --> 00:10:49,000 Speaker 1: So what would where would you go if you wanted 199 00:10:49,040 --> 00:10:52,120 Speaker 1: to have like a live call in show, for example, 200 00:10:52,760 --> 00:10:55,040 Speaker 1: and you were using maybe Skype for people to call 201 00:10:55,120 --> 00:10:57,960 Speaker 1: in so that you could answer questions live on the air, 202 00:10:58,120 --> 00:11:01,200 Speaker 1: You didn't really have that option. But because people couldn't 203 00:11:01,280 --> 00:11:03,480 Speaker 1: hear you and then respond to you in real time, 204 00:11:03,920 --> 00:11:07,079 Speaker 1: twitch or not twitch. But justin TV was a chance 205 00:11:07,120 --> 00:11:09,720 Speaker 1: of doing that. So it's kind of interesting. They said 206 00:11:09,760 --> 00:11:11,640 Speaker 1: that the only reason they were able to do it 207 00:11:11,720 --> 00:11:16,440 Speaker 1: is because while they had been developing justin dot tv, uh, 208 00:11:16,480 --> 00:11:18,640 Speaker 1: they had advanced the technology to a point that brought 209 00:11:18,720 --> 00:11:22,400 Speaker 1: the cost of streaming video down significantly to just like 210 00:11:22,480 --> 00:11:24,680 Speaker 1: a couple of cents or a couple of fractions of 211 00:11:24,679 --> 00:11:29,240 Speaker 1: assent UH for a minute of video, which was important 212 00:11:29,240 --> 00:11:32,079 Speaker 1: because you know, depending upon how much bandwidth you were using, 213 00:11:32,080 --> 00:11:34,120 Speaker 1: your I s P might be charging you quite a 214 00:11:34,120 --> 00:11:36,760 Speaker 1: bit for it. Oh sure, but at that rate, with 215 00:11:36,880 --> 00:11:42,640 Speaker 1: a combination of advertising revenue from from ads and also 216 00:11:42,760 --> 00:11:45,520 Speaker 1: money from investors, they could kind of plug along, yeah exactly. 217 00:11:45,559 --> 00:11:49,160 Speaker 1: The ad supported videos helped them recapture some of the costs, 218 00:11:49,160 --> 00:11:52,960 Speaker 1: and they also got some pretty significant funding, like their 219 00:11:53,040 --> 00:11:56,599 Speaker 1: their first round UH netted them about two million dollars. 220 00:11:56,640 --> 00:11:59,120 Speaker 1: They would later get I think twenty million a couple 221 00:11:59,120 --> 00:12:02,320 Speaker 1: of years down the road. So they were getting money 222 00:12:02,400 --> 00:12:05,880 Speaker 1: from investors who said, this is the future of this 223 00:12:06,000 --> 00:12:09,559 Speaker 1: kind of internet delivery system. Plus they were thinking ahead 224 00:12:09,600 --> 00:12:12,240 Speaker 1: of how can we monetize this so that once once 225 00:12:12,280 --> 00:12:15,240 Speaker 1: we get past that investment period, how do we keep 226 00:12:15,240 --> 00:12:18,640 Speaker 1: making money? And that was through the ad support, So 227 00:12:19,160 --> 00:12:21,960 Speaker 1: they began to think, how can we really make it 228 00:12:22,000 --> 00:12:24,160 Speaker 1: easier for people to find the stuff they want? Because 229 00:12:24,200 --> 00:12:27,400 Speaker 1: as more people join, it gets crowded, right, and you're like, well, 230 00:12:27,559 --> 00:12:29,400 Speaker 1: there's a ton of people on here, but I'm only 231 00:12:29,440 --> 00:12:33,600 Speaker 1: interested in one specific topic. Uh. And so they started 232 00:12:33,600 --> 00:12:36,440 Speaker 1: to make categories and things like people who wanted to 233 00:12:36,480 --> 00:12:39,600 Speaker 1: talk about business, or people who wanted to talk about fashion, 234 00:12:39,720 --> 00:12:43,000 Speaker 1: or people wanted to talk about sports or video games, 235 00:12:43,320 --> 00:12:47,120 Speaker 1: and the video games category began to get incredibly popular, 236 00:12:47,240 --> 00:12:50,880 Speaker 1: something that mystified justin by the way you can watch 237 00:12:50,920 --> 00:12:53,120 Speaker 1: video interviews, and he said, I just didn't get it. 238 00:12:53,160 --> 00:12:57,200 Speaker 1: People were playing video games and streaming their video game 239 00:12:57,280 --> 00:13:00,520 Speaker 1: experience online and other people were watching it. What's up 240 00:13:00,520 --> 00:13:03,600 Speaker 1: with that? He just he was like, I didn't understand it, 241 00:13:03,640 --> 00:13:06,320 Speaker 1: but I realized that other people found it valuable, So 242 00:13:06,360 --> 00:13:08,440 Speaker 1: we knew that it was something we had to look at. 243 00:13:09,240 --> 00:13:11,559 Speaker 1: So they start thinking, well, you know, this makes a 244 00:13:11,640 --> 00:13:13,679 Speaker 1: lot of sense. We've we've got all these people who 245 00:13:13,720 --> 00:13:17,960 Speaker 1: are um watching stuff. We should try and take advantage 246 00:13:17,960 --> 00:13:23,160 Speaker 1: of that. So they ultimately on June six, two eleven, 247 00:13:23,280 --> 00:13:26,959 Speaker 1: launched twitch dot tv in public beta as a completely 248 00:13:27,080 --> 00:13:30,120 Speaker 1: separate brand. Now, this was the idea of let's take 249 00:13:30,160 --> 00:13:33,960 Speaker 1: that video game content that everyone is streaming and create 250 00:13:34,000 --> 00:13:37,320 Speaker 1: a product specifically for that, with tools built in to 251 00:13:37,440 --> 00:13:40,240 Speaker 1: help people do that more smoothly, so they don't have 252 00:13:40,320 --> 00:13:43,600 Speaker 1: to create their own system at home to make it work. 253 00:13:43,640 --> 00:13:45,800 Speaker 1: We're going to try and integrate it in as many 254 00:13:45,840 --> 00:13:49,640 Speaker 1: different approaches as possible so that more people can do this, 255 00:13:49,679 --> 00:13:53,320 Speaker 1: because it's clearly a runaway success story, and it made 256 00:13:53,400 --> 00:13:55,679 Speaker 1: much more sense to create a new brand than to 257 00:13:55,800 --> 00:13:58,280 Speaker 1: just nestle it under justin dot tv. In fact, Twitch 258 00:13:58,320 --> 00:14:01,439 Speaker 1: dot tv was getting bigger than just dot tv. Yeah, 259 00:14:01,520 --> 00:14:04,760 Speaker 1: they they basically immediately found an audience. They started with 260 00:14:04,840 --> 00:14:09,320 Speaker 1: some three point two million unique visitors per month. Yep. Yeah, 261 00:14:09,440 --> 00:14:13,560 Speaker 1: they they've increased that a little bit. Yes, about a 262 00:14:13,679 --> 00:14:17,520 Speaker 1: year later, they would hit twenty million unique visitors per month. Yeah, 263 00:14:17,600 --> 00:14:20,520 Speaker 1: and I think they're at forty five million currently, so 264 00:14:20,720 --> 00:14:24,760 Speaker 1: it is essentially doubled, like year over year, more than 265 00:14:24,800 --> 00:14:27,239 Speaker 1: doubled on that first one. So that was pretty impressive. 266 00:14:27,280 --> 00:14:29,400 Speaker 1: And you know, keep in mind that back in June 267 00:14:29,400 --> 00:14:31,560 Speaker 1: two th eleven, it was just in beta. It wasn't 268 00:14:31,560 --> 00:14:36,120 Speaker 1: even the full release yet. So they started having lots 269 00:14:36,120 --> 00:14:39,239 Speaker 1: of Like there's certain games that that definitely drove participation, 270 00:14:39,320 --> 00:14:41,520 Speaker 1: StarCraft too being a big one. Anything that you can 271 00:14:41,520 --> 00:14:45,120 Speaker 1: get big tournaments going for exactly. So, and you know 272 00:14:45,160 --> 00:14:49,200 Speaker 1: there are places like in South Korea where StarCraft tournaments are, 273 00:14:49,520 --> 00:14:52,400 Speaker 1: like the super Bowl is here in the United States, 274 00:14:52,720 --> 00:14:56,040 Speaker 1: so you get these enormous crowds live audiences watching these 275 00:14:56,280 --> 00:14:58,640 Speaker 1: well only here in the US. It just it never 276 00:14:58,720 --> 00:15:02,880 Speaker 1: really caught on that way, mostly because production costs or 277 00:15:02,920 --> 00:15:05,480 Speaker 1: something like that. If you wanted to televise it, we're 278 00:15:05,560 --> 00:15:09,520 Speaker 1: so high that you would have trouble getting the audience 279 00:15:09,600 --> 00:15:13,080 Speaker 1: large enough to justify that expense. But the overhead for 280 00:15:13,160 --> 00:15:16,840 Speaker 1: Twitch is way lower. I mean, you don't have as 281 00:15:16,880 --> 00:15:21,960 Speaker 1: many employees for one thing, the infrastructure is once you've 282 00:15:22,200 --> 00:15:26,520 Speaker 1: made the investment, much less expensive than television. So it 283 00:15:26,600 --> 00:15:29,280 Speaker 1: was sort of the perfect platform for that and it 284 00:15:29,360 --> 00:15:32,720 Speaker 1: really took off. It also allows the audience to talk 285 00:15:32,720 --> 00:15:35,080 Speaker 1: to each other during the event in a way that 286 00:15:35,240 --> 00:15:39,000 Speaker 1: the normal broadcast television obviously doesn't allow unless you invite 287 00:15:39,040 --> 00:15:41,400 Speaker 1: a lot of your friends over right, or you have 288 00:15:41,720 --> 00:15:43,880 Speaker 1: some app where you can watch the show and talk 289 00:15:43,920 --> 00:15:46,120 Speaker 1: with other fans. But we've seen this over and over 290 00:15:46,200 --> 00:15:49,840 Speaker 1: that television makes it. It's really hard to get something 291 00:15:49,880 --> 00:15:52,920 Speaker 1: like that to work from a television platform and poured 292 00:15:52,920 --> 00:15:56,920 Speaker 1: it over into an online interactive platform. Usually you see 293 00:15:56,920 --> 00:15:59,000 Speaker 1: people just use either Twitter or Facebook for that kind 294 00:15:59,000 --> 00:16:01,800 Speaker 1: of thing, right, And in eleven that wasn't so much 295 00:16:01,840 --> 00:16:04,160 Speaker 1: the case quite yet, No, no, you know, so this 296 00:16:04,320 --> 00:16:06,280 Speaker 1: in this case, you have a platform that by its 297 00:16:06,360 --> 00:16:09,960 Speaker 1: very nature allows for that kind of thing. However, as 298 00:16:09,960 --> 00:16:13,520 Speaker 1: of eleven it's still was not profitable and would not 299 00:16:13,640 --> 00:16:17,160 Speaker 1: become I think it technically has never been profitable for 300 00:16:17,240 --> 00:16:23,000 Speaker 1: a full year. They are estimating that as it will be, right, 301 00:16:23,160 --> 00:16:25,680 Speaker 1: But well, well we'll see. You have to get that 302 00:16:25,880 --> 00:16:30,840 Speaker 1: great balance of AD dollars and user base in order 303 00:16:30,880 --> 00:16:33,880 Speaker 1: to make it profitable. And of course the bigger it gets, 304 00:16:34,080 --> 00:16:35,760 Speaker 1: the more of an investment you have to make in 305 00:16:35,800 --> 00:16:39,880 Speaker 1: the company's infrastructure. Right. Of course, us a few things 306 00:16:39,960 --> 00:16:43,640 Speaker 1: that helped out greatly with that. Yeah. Yeah. For one thing, 307 00:16:43,920 --> 00:16:47,760 Speaker 1: you've got the fact that both Sony and Microsoft gave 308 00:16:47,800 --> 00:16:52,120 Speaker 1: Twitch a big thumbs up. Right, they announced partnerships that 309 00:16:52,160 --> 00:16:56,080 Speaker 1: would allow a native app on both of those gaming 310 00:16:56,080 --> 00:16:59,760 Speaker 1: platforms to integrate Twitch tv. Right. So, with the case 311 00:16:59,800 --> 00:17:02,560 Speaker 1: of of PlayStation four, there's a button on the controller 312 00:17:02,560 --> 00:17:04,840 Speaker 1: where you can start sharing what you're playing. It's a 313 00:17:04,880 --> 00:17:07,560 Speaker 1: little more complex with Xbox one. Uh. And in both 314 00:17:07,600 --> 00:17:10,800 Speaker 1: cases it was a feature that was not immediately implemented 315 00:17:10,800 --> 00:17:13,399 Speaker 1: as soon as the consoles came out, but was implemented 316 00:17:13,440 --> 00:17:16,639 Speaker 1: a little bit later on the life cycle of those consoles, 317 00:17:17,119 --> 00:17:19,000 Speaker 1: I say life cycle, and these things haven't even been 318 00:17:19,000 --> 00:17:21,639 Speaker 1: out a full year yet. Yeah, it's really interesting to 319 00:17:21,720 --> 00:17:24,720 Speaker 1: see that they were throwing their support behind this because 320 00:17:24,760 --> 00:17:28,160 Speaker 1: they could see both Sony and Microsoft could see that 321 00:17:28,480 --> 00:17:30,960 Speaker 1: this this was a trend that was growing, that people 322 00:17:31,160 --> 00:17:34,360 Speaker 1: were eager to share their experiences and more people were 323 00:17:34,400 --> 00:17:37,680 Speaker 1: eager to to eat it all up. So I mean, 324 00:17:37,760 --> 00:17:40,639 Speaker 1: I've watched play throughs online. Of course, the playthroughs I 325 00:17:40,640 --> 00:17:43,680 Speaker 1: watched tend to be done for for comedic purposes, where 326 00:17:43,680 --> 00:17:45,679 Speaker 1: it's a bunch of guys who are playing who know 327 00:17:45,760 --> 00:17:48,440 Speaker 1: each other really well, and they're really doing it as 328 00:17:48,520 --> 00:17:51,520 Speaker 1: a way of entertaining other people, Like that's the whole 329 00:17:51,520 --> 00:17:55,040 Speaker 1: purpose of it. They're not playing a game competitively against others, 330 00:17:55,040 --> 00:17:57,520 Speaker 1: which is what a lot of the Twitch TV stuffs for. 331 00:17:58,200 --> 00:18:01,359 Speaker 1: Not everything, but also you have you have streamed yourself 332 00:18:01,440 --> 00:18:04,160 Speaker 1: on Twitch, haven't you? I did. I did for twenty 333 00:18:04,160 --> 00:18:08,560 Speaker 1: five hours straight. I did a stream of me just 334 00:18:08,760 --> 00:18:12,440 Speaker 1: playing Minecraft. It was for the Extra Life charity, which 335 00:18:12,520 --> 00:18:15,680 Speaker 1: raises money for to buy video games and video game 336 00:18:15,720 --> 00:18:19,720 Speaker 1: systems for kids who are in hospitals and uh, so 337 00:18:19,840 --> 00:18:21,199 Speaker 1: I was part of this. It was kind of like 338 00:18:21,200 --> 00:18:24,679 Speaker 1: a Pledge Drive type thing. And for twenty five hours 339 00:18:24,680 --> 00:18:27,959 Speaker 1: I played and and streamed it live, and I always 340 00:18:28,000 --> 00:18:30,480 Speaker 1: had at least one or two people watching. Sometimes I 341 00:18:30,520 --> 00:18:33,240 Speaker 1: would have you know, a few dozen watching at a time. 342 00:18:33,920 --> 00:18:36,240 Speaker 1: UH And I even had a thing where for certain 343 00:18:36,359 --> 00:18:41,800 Speaker 1: levels of contributions, I would build uh monuments within Minecraft, 344 00:18:42,000 --> 00:18:44,879 Speaker 1: and the more money you made, the bigger and more 345 00:18:45,040 --> 00:18:48,119 Speaker 1: rare the monument would be, like the materials would be 346 00:18:48,119 --> 00:18:53,080 Speaker 1: more rare. So some people were incredibly generous and I 347 00:18:53,160 --> 00:18:56,640 Speaker 1: had to go and find the rarest materials and Minecraft 348 00:18:56,880 --> 00:18:59,399 Speaker 1: within twenty five hours and still build it. Like actually, 349 00:18:59,400 --> 00:19:01,320 Speaker 1: I was like, all, so for the first eighteen hours, 350 00:19:01,400 --> 00:19:04,959 Speaker 1: I'm just mining and then for the last seven hours 351 00:19:04,960 --> 00:19:07,880 Speaker 1: I'll be building. So yeah, it was actually a lot 352 00:19:07,920 --> 00:19:10,080 Speaker 1: of fun. But because of Twitch, I was able to 353 00:19:10,080 --> 00:19:12,200 Speaker 1: stream it live and people could check in on me 354 00:19:12,240 --> 00:19:14,600 Speaker 1: and see that I was actually doing this thing that 355 00:19:14,640 --> 00:19:17,200 Speaker 1: I had said I would do. UM And I also, 356 00:19:17,280 --> 00:19:19,719 Speaker 1: you know, I got to have an ongoing conversation with 357 00:19:19,760 --> 00:19:22,159 Speaker 1: folks through chat. He would chat to me and I 358 00:19:22,160 --> 00:19:24,680 Speaker 1: would answer that questions live while playing and and kind 359 00:19:24,680 --> 00:19:27,639 Speaker 1: of keep you motivated while you were doing this ridiculous marathon. 360 00:19:27,800 --> 00:19:30,720 Speaker 1: Sometime around our number nineteen or twenty, I definitely needed 361 00:19:30,720 --> 00:19:33,200 Speaker 1: some more motivation. Not to mention coffee. I think that's 362 00:19:33,240 --> 00:19:35,640 Speaker 1: the wall that they talked about in Marathon ng UM. 363 00:19:35,640 --> 00:19:39,639 Speaker 1: In September, Twitch sold an undisclosed stake in the company 364 00:19:39,720 --> 00:19:43,080 Speaker 1: to investors, raising that twenty million dollars that you mentioned earlier, 365 00:19:43,119 --> 00:19:45,760 Speaker 1: and a lot of that went to supporting their rollout 366 00:19:45,800 --> 00:19:48,479 Speaker 1: into the PS four and Xbox one. Yeah. Now, if 367 00:19:48,480 --> 00:19:51,800 Speaker 1: you're looking at today's numbers, according to Mercury News that 368 00:19:51,920 --> 00:19:56,359 Speaker 1: this was astounding. Twitch dot tv accounts forty percent of 369 00:19:56,520 --> 00:20:01,080 Speaker 1: all United States live streaming online. That's more than any 370 00:20:01,080 --> 00:20:06,159 Speaker 1: other live streaming site, including ESPN, NBC Sports and YouTube Live. Um. 371 00:20:06,400 --> 00:20:08,600 Speaker 1: This is all according to Quilt, which is a company 372 00:20:08,640 --> 00:20:12,360 Speaker 1: that helps networks manage bandwidth demand for video online and 373 00:20:12,440 --> 00:20:15,159 Speaker 1: announced that they had begun working with Twitch in April. 374 00:20:16,520 --> 00:20:21,280 Speaker 1: Twitch tv receives more than forty million unique visitors each month, 375 00:20:21,359 --> 00:20:24,440 Speaker 1: which I think we mentioned before. But yeah, that's a lot. 376 00:20:25,000 --> 00:20:28,359 Speaker 1: Like sure, if you look at YouTube, YouTube is getting 377 00:20:28,359 --> 00:20:31,720 Speaker 1: a billion, okay, but YouTube has Google behind it and 378 00:20:31,840 --> 00:20:35,399 Speaker 1: is adding a hundred hours of footage every minute, right, 379 00:20:35,840 --> 00:20:37,919 Speaker 1: This is this is considering that this is something that 380 00:20:38,000 --> 00:20:40,960 Speaker 1: was a spinoff of another company and has only been 381 00:20:41,440 --> 00:20:45,920 Speaker 1: around for for less than four full years, it's phenomenal. Yeah, 382 00:20:45,920 --> 00:20:47,879 Speaker 1: and its users spend an average of a hundred and 383 00:20:47,960 --> 00:20:52,840 Speaker 1: six minutes per day watching streams average. That is also insane. 384 00:20:52,920 --> 00:20:56,040 Speaker 1: That's so not to spoil it, but that's going to 385 00:20:56,160 --> 00:20:57,960 Speaker 1: come into play when we have our conversation of why 386 00:20:57,960 --> 00:20:59,800 Speaker 1: the heck would Google wants to spend a billion dollars 387 00:20:59,800 --> 00:21:01,760 Speaker 1: on the yes. To put all of this in perspective 388 00:21:01,800 --> 00:21:05,040 Speaker 1: a little bit, In February, a company called deep Field, 389 00:21:05,080 --> 00:21:07,919 Speaker 1: which does Internet data analysis for media companies like the 390 00:21:07,920 --> 00:21:11,280 Speaker 1: Wall Street Journal, Ransom numbers and reported that Twitch accounts 391 00:21:11,320 --> 00:21:14,640 Speaker 1: for one point eight percent of United States peak internet traffic, 392 00:21:14,880 --> 00:21:18,359 Speaker 1: which places it fourth overall after Netflix, which stands at 393 00:21:19,040 --> 00:21:24,840 Speaker 1: Google and Apple four point three. That also puts it 394 00:21:24,920 --> 00:21:28,359 Speaker 1: ahead of companies like Hulu, Facebook, and Valve, all of 395 00:21:28,359 --> 00:21:31,280 Speaker 1: which come in a couple tents of a percentage point lower. 396 00:21:31,480 --> 00:21:35,120 Speaker 1: That's pretty amazing. I mean, now, granted, these are all 397 00:21:35,200 --> 00:21:39,480 Speaker 1: different things, right because with Google, the fact that is 398 00:21:39,520 --> 00:21:43,160 Speaker 1: Google is pretty impressive considering that that's it's text based. Yeah, yeah, 399 00:21:43,160 --> 00:21:46,199 Speaker 1: it's not just uh, I mean, I guess Google that 400 00:21:46,200 --> 00:21:49,359 Speaker 1: probably also includes YouTube, so that's probably some video in 401 00:21:49,400 --> 00:21:52,199 Speaker 1: there too. But yeah, it's still impressive. The fact that 402 00:21:52,600 --> 00:21:55,240 Speaker 1: Valve is so far back is also impressive because you're 403 00:21:55,280 --> 00:21:59,680 Speaker 1: talking about like not just uh, online gaming, but that's 404 00:21:59,760 --> 00:22:02,320 Speaker 1: a gaming delivery service, you know, if people are buying 405 00:22:02,359 --> 00:22:05,800 Speaker 1: games on Steam. But um, yeah, still pretty impressive. Well, 406 00:22:06,200 --> 00:22:11,320 Speaker 1: we're gonna next tackle exactly how this whole system works, 407 00:22:11,359 --> 00:22:13,720 Speaker 1: but before we do that, let's take a quick break 408 00:22:13,880 --> 00:22:17,040 Speaker 1: to thank our sponsor. Alright, so we're back. Let's talk 409 00:22:17,080 --> 00:22:21,040 Speaker 1: about how this company actually delivers live streaming. That was 410 00:22:21,119 --> 00:22:24,720 Speaker 1: Daniel's question, how does it work? Now? A lot of 411 00:22:24,760 --> 00:22:27,760 Speaker 1: live streaming is using a very similar method, But first 412 00:22:27,920 --> 00:22:30,919 Speaker 1: it's good to say what it does not do, right, 413 00:22:30,960 --> 00:22:34,840 Speaker 1: because it's an intrinsically different system than for example, YouTube. Yeah, 414 00:22:34,840 --> 00:22:38,119 Speaker 1: so YouTube, you've got prerecorded user videos for the most part. 415 00:22:38,200 --> 00:22:40,760 Speaker 1: I mean, YouTube has some live stuff now, but the 416 00:22:40,760 --> 00:22:44,600 Speaker 1: bulk of its business is prerecorded video. So with that, 417 00:22:44,960 --> 00:22:47,320 Speaker 1: I would, for example, if I were uploading a video, 418 00:22:47,400 --> 00:22:50,280 Speaker 1: it would upload to YouTube, then it processes and it's 419 00:22:50,320 --> 00:22:55,920 Speaker 1: stored on YouTube servers actually usually multiple servers. Redundancy. Hey, 420 00:22:56,040 --> 00:22:58,000 Speaker 1: we said at the same time, we were being redundant. 421 00:22:58,440 --> 00:23:02,240 Speaker 1: So we're trying to it serve it up from a disk. 422 00:23:02,480 --> 00:23:05,560 Speaker 1: You know. That's YouTube actually pulls the video from some 423 00:23:05,720 --> 00:23:10,119 Speaker 1: saved part on some server or servers and serves it 424 00:23:10,240 --> 00:23:12,000 Speaker 1: up to you when you say I want to watch 425 00:23:12,040 --> 00:23:15,880 Speaker 1: this video now. That is completely different from the way 426 00:23:15,960 --> 00:23:18,800 Speaker 1: live streaming works. And one of the things that made 427 00:23:18,840 --> 00:23:23,399 Speaker 1: it possible was that, uh, this idea of peering networks. 428 00:23:23,720 --> 00:23:26,080 Speaker 1: So I'll get to that, but but also to explain 429 00:23:26,160 --> 00:23:30,840 Speaker 1: how peering networks are different, your basic uh server client 430 00:23:30,920 --> 00:23:34,800 Speaker 1: relationship is very centralized. You've got a server that has 431 00:23:34,840 --> 00:23:37,439 Speaker 1: all the information on it, and then you use a 432 00:23:37,480 --> 00:23:40,639 Speaker 1: client like a browser that then sends a query to 433 00:23:40,720 --> 00:23:42,720 Speaker 1: the server which says, oh, I know what you want, 434 00:23:42,760 --> 00:23:46,000 Speaker 1: you want this and sends it to you. Centralized right right, 435 00:23:46,040 --> 00:23:48,120 Speaker 1: And that means that if too many people are trying 436 00:23:48,119 --> 00:23:50,560 Speaker 1: to access that server at the same time, you're going 437 00:23:50,600 --> 00:23:53,560 Speaker 1: to run into a little bit of lags exactly, which 438 00:23:53,600 --> 00:23:55,680 Speaker 1: comes down on to your end because the server is 439 00:23:55,680 --> 00:23:57,680 Speaker 1: going to say, oh, hey, wait a minute, I'm doing 440 00:23:57,680 --> 00:23:59,520 Speaker 1: this other thing and this other thing, and and it 441 00:23:59,640 --> 00:24:03,320 Speaker 1: resolves and basically a load bar exactly. Yeah, the same 442 00:24:03,359 --> 00:24:06,919 Speaker 1: sort of thing. Like if Lauren had, uh, you know 443 00:24:07,240 --> 00:24:09,080 Speaker 1: she has she has cookies at her desk, and the 444 00:24:09,280 --> 00:24:11,879 Speaker 1: entire office comes to her at the same time to 445 00:24:11,920 --> 00:24:15,280 Speaker 1: get a cookie each It's gonna take a while because, 446 00:24:15,880 --> 00:24:17,600 Speaker 1: you know, it may be that she has a variety 447 00:24:17,640 --> 00:24:19,920 Speaker 1: of cookies and we each want a different type of cookies. 448 00:24:19,920 --> 00:24:22,880 Speaker 1: So Lauren, being Lauren, wants to be fair to everybody 449 00:24:23,000 --> 00:24:25,440 Speaker 1: and tries to make sure that this happens. But Lauren, 450 00:24:25,600 --> 00:24:28,679 Speaker 1: also being Lauren, gets overwhelmed with the fact that a 451 00:24:28,680 --> 00:24:32,000 Speaker 1: whole bunch of cookie hungry maniacs are approaching her and 452 00:24:32,160 --> 00:24:35,240 Speaker 1: wigs out. I mean, it just happens. So that would 453 00:24:35,240 --> 00:24:37,320 Speaker 1: be an issue of client server. So how can you 454 00:24:37,320 --> 00:24:39,639 Speaker 1: get around that. Well, if Lauren was able to clone 455 00:24:39,640 --> 00:24:43,200 Speaker 1: herself and every single version of Lauren had a box 456 00:24:43,200 --> 00:24:45,439 Speaker 1: of cookies, and all of the box of cookies had 457 00:24:45,480 --> 00:24:48,400 Speaker 1: identical types of cookies in them, like the whole variety, 458 00:24:48,720 --> 00:24:51,120 Speaker 1: we could all go to different Lauren's to get our cookies. 459 00:24:51,440 --> 00:24:54,080 Speaker 1: The interesting thing is that you know, they'd all have 460 00:24:54,200 --> 00:24:58,080 Speaker 1: exactly the cookie we would want. I want a cookie, 461 00:24:58,320 --> 00:25:01,520 Speaker 1: I like this bizarre university. Well that's the that's kind 462 00:25:01,560 --> 00:25:04,560 Speaker 1: of the nature of appearing. So the other issue here 463 00:25:04,920 --> 00:25:06,959 Speaker 1: is that if you have the prerecorded material, you can 464 00:25:07,000 --> 00:25:10,800 Speaker 1: do something called over subscribing bandwidth. Now, that just means 465 00:25:10,840 --> 00:25:13,240 Speaker 1: that you can serve more data than the network can 466 00:25:13,240 --> 00:25:16,240 Speaker 1: actually accommodate at any given time. So let's say that 467 00:25:16,280 --> 00:25:19,119 Speaker 1: I've got a five gigabyte limit on how much data 468 00:25:19,160 --> 00:25:21,680 Speaker 1: i can push out at any given time, but I've 469 00:25:21,680 --> 00:25:25,000 Speaker 1: got seven gigabytes of demand, all right. The way that 470 00:25:25,080 --> 00:25:28,280 Speaker 1: over subscription works is that I send out messages that 471 00:25:28,359 --> 00:25:30,719 Speaker 1: tell all the different clients that are trying to access 472 00:25:30,800 --> 00:25:33,760 Speaker 1: my stuff. Hey, I'm getting to you. Here's some of 473 00:25:33,800 --> 00:25:37,840 Speaker 1: the information. More of its coming right now, but just 474 00:25:37,840 --> 00:25:39,760 Speaker 1: just you know what, do a little song and dance 475 00:25:39,760 --> 00:25:42,840 Speaker 1: for the user. That's when you get the little buffer thing. Right. 476 00:25:42,960 --> 00:25:45,679 Speaker 1: So with YouTube, if a video is buffering mainly, that 477 00:25:45,800 --> 00:25:49,040 Speaker 1: just means that the service has a lot of traffic 478 00:25:49,080 --> 00:25:51,080 Speaker 1: and it's trying to serve all of that traffic at 479 00:25:51,080 --> 00:25:53,720 Speaker 1: the same time. So you're getting bits and bits of 480 00:25:53,800 --> 00:25:56,080 Speaker 1: data that are buffering slowly, so that once you hit 481 00:25:56,119 --> 00:25:59,040 Speaker 1: a certain threshold, you can push play and watch your 482 00:25:59,119 --> 00:26:03,480 Speaker 1: video and everything goes well, yeah, exactly. Hypothetically, exactly. Yeah, 483 00:26:03,520 --> 00:26:06,240 Speaker 1: if if, if it has buffered enough, you can push 484 00:26:06,240 --> 00:26:08,240 Speaker 1: play and you can watch the video all the way 485 00:26:08,280 --> 00:26:11,879 Speaker 1: through because at that point you're getting data faster than 486 00:26:11,920 --> 00:26:14,760 Speaker 1: the video is completing. Now, anyone who's ever seen video 487 00:26:14,760 --> 00:26:16,520 Speaker 1: buffer in the middle of a video know that that 488 00:26:16,560 --> 00:26:19,160 Speaker 1: knows that that doesn't always work, you know. Sometimes it's 489 00:26:19,160 --> 00:26:21,720 Speaker 1: just one of those things where the playback catches up 490 00:26:22,040 --> 00:26:23,760 Speaker 1: to the amount of data that's come in and you 491 00:26:23,760 --> 00:26:27,159 Speaker 1: have to wait longer. Okay, so how is how is 492 00:26:27,200 --> 00:26:32,000 Speaker 1: this peering structure different? Alright? So it's decentralized. So instead 493 00:26:32,000 --> 00:26:35,439 Speaker 1: of having that centralized server that all the different clients 494 00:26:35,440 --> 00:26:40,040 Speaker 1: are contacting, you have a system, a network architecture where 495 00:26:40,320 --> 00:26:44,120 Speaker 1: multiple servers are using copies of the same sort of thing, 496 00:26:44,480 --> 00:26:46,679 Speaker 1: and all of them can take on a part of 497 00:26:46,720 --> 00:26:49,879 Speaker 1: the work. So instead of one server where you you know, 498 00:26:49,920 --> 00:26:52,000 Speaker 1: you're just souping it up as fast as it can 499 00:26:52,000 --> 00:26:54,520 Speaker 1: go so it can work as best as it possibly can, 500 00:26:55,040 --> 00:26:59,120 Speaker 1: you distribute it across an entire network. Google actually does 501 00:26:59,119 --> 00:27:02,640 Speaker 1: this for their search, So for a Google Search query, 502 00:27:02,720 --> 00:27:05,439 Speaker 1: they have an entire network of servers that work together. 503 00:27:05,560 --> 00:27:08,080 Speaker 1: That's why if you ever do that Google search and 504 00:27:08,080 --> 00:27:10,480 Speaker 1: you look and it says it took point zero zero 505 00:27:10,560 --> 00:27:14,920 Speaker 1: zero zero zero eight three seconds to to answer this query. 506 00:27:15,000 --> 00:27:17,919 Speaker 1: That's why, because I've got this distributed network, So just 507 00:27:17,920 --> 00:27:20,359 Speaker 1: think of it as a group of people all taking 508 00:27:20,480 --> 00:27:26,240 Speaker 1: some of the load so that no one server is overworked. Now, 509 00:27:26,280 --> 00:27:29,320 Speaker 1: these servers can be located in different data centers, which 510 00:27:29,320 --> 00:27:31,720 Speaker 1: could be all the way across the world. They don't 511 00:27:31,840 --> 00:27:34,560 Speaker 1: have to be co located with anything or next to 512 00:27:34,600 --> 00:27:39,000 Speaker 1: each other. And because of this, you've got these individual 513 00:27:39,080 --> 00:27:42,760 Speaker 1: nodes which have this peering architecture that they're they're the peers, 514 00:27:43,160 --> 00:27:45,480 Speaker 1: so that each individual server is a peer in this 515 00:27:45,560 --> 00:27:49,119 Speaker 1: peering architecture, and they act as both suppliers and consumers 516 00:27:49,119 --> 00:27:52,119 Speaker 1: of resources. So each peer makes part of its computer 517 00:27:52,160 --> 00:27:54,720 Speaker 1: resources available to the other peers. This is that sharing, 518 00:27:54,960 --> 00:27:58,520 Speaker 1: that distribution of the load of any given computer request. 519 00:27:58,960 --> 00:28:01,399 Speaker 1: So this way you can get a bunch of you know, 520 00:28:01,480 --> 00:28:05,000 Speaker 1: average servers together and network them together and get a 521 00:28:05,000 --> 00:28:07,760 Speaker 1: better performance than if you just invest in that one 522 00:28:07,800 --> 00:28:11,600 Speaker 1: super fast server. It also makes it cheaper because you 523 00:28:11,640 --> 00:28:13,520 Speaker 1: can just get off the shelf parts to build your 524 00:28:13,560 --> 00:28:17,520 Speaker 1: servers instead of buying you know, the next ultra fast exactly. 525 00:28:17,880 --> 00:28:19,879 Speaker 1: So it and Again, this is also what allows you 526 00:28:19,920 --> 00:28:22,840 Speaker 1: to be have a lot of redundancy because you can 527 00:28:22,920 --> 00:28:26,680 Speaker 1: you can get a bunch of inexpensive parts as opposed to, well, 528 00:28:26,800 --> 00:28:29,200 Speaker 1: we blew all our money on this one really fast 529 00:28:29,240 --> 00:28:32,320 Speaker 1: machine that can't live up to our expectations. So what 530 00:28:32,480 --> 00:28:34,760 Speaker 1: happens I mean, because I'm sure that with the number 531 00:28:34,760 --> 00:28:37,160 Speaker 1: of people who are using this service, what happens if 532 00:28:37,320 --> 00:28:42,040 Speaker 1: demand exceeds even this capacity? Okay, so if there were 533 00:28:42,080 --> 00:28:46,200 Speaker 1: no overflow method, you would all get a buffering message 534 00:28:46,200 --> 00:28:49,120 Speaker 1: at the same time. Because all of these servers are 535 00:28:49,120 --> 00:28:51,400 Speaker 1: working together. It's this whole It's like it's an all 536 00:28:51,480 --> 00:28:55,680 Speaker 1: or nothing kind of situation. Now. Uh, the the peak 537 00:28:55,880 --> 00:28:59,920 Speaker 1: use of Twitch was something like ten times the amount 538 00:29:00,080 --> 00:29:02,560 Speaker 1: of the average use, which meant that they had to 539 00:29:02,600 --> 00:29:06,680 Speaker 1: build a network that could withstand ten times the normal 540 00:29:06,800 --> 00:29:08,719 Speaker 1: use at least, and in fact, you want to have 541 00:29:08,800 --> 00:29:11,960 Speaker 1: more than that to have a comfortable buffer. But even 542 00:29:12,000 --> 00:29:14,280 Speaker 1: if it goes beyond that. This is something they've learned 543 00:29:14,280 --> 00:29:17,360 Speaker 1: from their just in TV days, they wanted the ability 544 00:29:17,400 --> 00:29:20,560 Speaker 1: to have an overflow capacity where they use something called 545 00:29:20,560 --> 00:29:24,280 Speaker 1: a content delivery network or c d N. Now, this 546 00:29:24,360 --> 00:29:27,880 Speaker 1: is a specific type of business. It's a distributed system 547 00:29:27,880 --> 00:29:30,880 Speaker 1: of servers, so again sort of appeering network, but this 548 00:29:30,920 --> 00:29:34,000 Speaker 1: is kind of appeering network for higher So you would 549 00:29:34,320 --> 00:29:36,880 Speaker 1: use this if you wanted to create content and you 550 00:29:36,880 --> 00:29:39,160 Speaker 1: want to be able to distribute it to a wide 551 00:29:39,200 --> 00:29:42,200 Speaker 1: audience on a live basis. This is the sort of 552 00:29:42,240 --> 00:29:44,200 Speaker 1: approach you would want to take. And this is what 553 00:29:44,680 --> 00:29:47,640 Speaker 1: some companies like Netflix, they depend upon the sort of 554 00:29:47,680 --> 00:29:51,160 Speaker 1: networks to deliver the stuff that they make. Some companies 555 00:29:51,200 --> 00:29:53,840 Speaker 1: invest in their own, like twitch tv has its own network, 556 00:29:53,880 --> 00:29:57,080 Speaker 1: but again, if it exceeds that capacity, they can then 557 00:29:57,160 --> 00:29:59,560 Speaker 1: say tap the shoulder of the c d N and 558 00:29:59,600 --> 00:30:03,240 Speaker 1: say can you take over some of this stuff. Now, 559 00:30:03,280 --> 00:30:06,040 Speaker 1: for these cases, it does mean that the company, whether 560 00:30:06,080 --> 00:30:09,040 Speaker 1: it's twitched TV or some other media company, has to 561 00:30:09,120 --> 00:30:11,680 Speaker 1: pay for those services. You know, it's not a free thing, 562 00:30:12,200 --> 00:30:14,760 Speaker 1: but it does mean that there's at least a safety 563 00:30:14,800 --> 00:30:18,200 Speaker 1: net there. Should Let's say that that there's some huge 564 00:30:18,240 --> 00:30:21,200 Speaker 1: event that happens on twitch tv that really gets the 565 00:30:21,360 --> 00:30:24,080 Speaker 1: entire gaming world interested and they all want to watch 566 00:30:24,160 --> 00:30:27,520 Speaker 1: it at the same time live streaming, then they might 567 00:30:27,560 --> 00:30:31,800 Speaker 1: need to use that kind of resource because the demand 568 00:30:32,080 --> 00:30:36,880 Speaker 1: would exceed the capacity. They also developed a load balancing 569 00:30:37,040 --> 00:30:40,080 Speaker 1: software called Usher to help out a little bit in 570 00:30:40,080 --> 00:30:43,040 Speaker 1: in high demand situations like that, Yeah, this is really 571 00:30:43,120 --> 00:30:47,880 Speaker 1: really cool stuff. So Usher is kind of the the 572 00:30:47,880 --> 00:30:52,240 Speaker 1: the traffic manager. Right. Usher identifies what requests are and 573 00:30:52,280 --> 00:30:54,960 Speaker 1: how to fulfill those requests in what in other words, 574 00:30:54,960 --> 00:30:57,560 Speaker 1: how how many servers it needs to call upon in 575 00:30:57,640 --> 00:31:00,080 Speaker 1: order to handle the demand that's coming in. Right, So, 576 00:31:00,160 --> 00:31:04,360 Speaker 1: let's say that I'm going to stream the playing uh, 577 00:31:04,480 --> 00:31:07,320 Speaker 1: I don't know, some game, like I'm gonna go back 578 00:31:07,360 --> 00:31:09,800 Speaker 1: and play through some of the Halo games and I'm 579 00:31:09,800 --> 00:31:12,880 Speaker 1: gonna stream them live. Now, in that case, there's probably 580 00:31:12,920 --> 00:31:14,880 Speaker 1: not going to be a huge demand. There aren't that 581 00:31:14,920 --> 00:31:17,520 Speaker 1: many people who are that interested in watching a bald 582 00:31:17,520 --> 00:31:22,520 Speaker 1: guy fail miserably at playing Halo. Um until you try, 583 00:31:22,600 --> 00:31:24,800 Speaker 1: that's true. But I'm assuming that this would not be 584 00:31:24,880 --> 00:31:27,400 Speaker 1: one that Usher would be like ho Alert the media. 585 00:31:27,840 --> 00:31:31,840 Speaker 1: So Usher, when getting the message that I'm going to 586 00:31:31,880 --> 00:31:36,160 Speaker 1: stream this video and then analyzing how much of the 587 00:31:36,240 --> 00:31:39,600 Speaker 1: resources available we're going to be necessary for to deliver 588 00:31:39,720 --> 00:31:42,800 Speaker 1: upon that would contact various servers to make sure that 589 00:31:42,840 --> 00:31:45,320 Speaker 1: it was covered. And usually we're talking about a capacity 590 00:31:45,360 --> 00:31:48,240 Speaker 1: that exceeds what it expects for the demand. Now, if 591 00:31:48,240 --> 00:31:52,720 Speaker 1: it were some uh star Craft two champions that never 592 00:31:52,760 --> 00:31:55,160 Speaker 1: faced each other before, and they were going to have 593 00:31:55,280 --> 00:31:58,360 Speaker 1: a tournament style game against each other to determine who 594 00:31:58,520 --> 00:32:02,560 Speaker 1: is the true champion StarCraft Too. That could be huge, 595 00:32:02,600 --> 00:32:05,600 Speaker 1: and Usher would then make that determination and contact the 596 00:32:05,680 --> 00:32:09,600 Speaker 1: various servers that would all be copying the stream so 597 00:32:09,680 --> 00:32:12,120 Speaker 1: that they could share it up to UH to anyone 598 00:32:12,200 --> 00:32:15,400 Speaker 1: that would wanted to watch it. So then you've got 599 00:32:16,160 --> 00:32:18,600 Speaker 1: all these different factors that come into play that Usher 600 00:32:18,640 --> 00:32:21,719 Speaker 1: will take into considerations, So not just how popular are 601 00:32:21,720 --> 00:32:23,520 Speaker 1: the stream is going to be, but how much is 602 00:32:23,520 --> 00:32:27,000 Speaker 1: it gonna cost to serve up the live video? Because 603 00:32:27,040 --> 00:32:30,480 Speaker 1: like I said, bandwidth is not necessarily free. And Usher 604 00:32:30,560 --> 00:32:32,200 Speaker 1: can sit there and look and say which one of 605 00:32:32,240 --> 00:32:34,680 Speaker 1: these options is going to be the cheapest. Now, in 606 00:32:34,720 --> 00:32:38,520 Speaker 1: some cases, like with that StarCraft Too example, cheapest may 607 00:32:38,560 --> 00:32:41,480 Speaker 1: not be the best response because they want to make 608 00:32:41,520 --> 00:32:45,800 Speaker 1: sure that their customers, the viewers, are satisfied. So they 609 00:32:45,880 --> 00:32:48,200 Speaker 1: might say, all right, you know what, We're gonna incur 610 00:32:48,320 --> 00:32:49,880 Speaker 1: a cost on this, but what we want to do 611 00:32:49,960 --> 00:32:52,600 Speaker 1: is make sure we have the lowest latency possible so 612 00:32:52,600 --> 00:32:55,600 Speaker 1: that people who are watching are not experiencing like a 613 00:32:55,720 --> 00:33:00,440 Speaker 1: thirty second lag behind the action, so then sure can 614 00:33:00,480 --> 00:33:05,240 Speaker 1: prioritize on the fly. And it apparently the controls here 615 00:33:05,360 --> 00:33:09,880 Speaker 1: are incredibly precise so that you can tweak exactly which 616 00:33:09,920 --> 00:33:13,160 Speaker 1: things are the most important for any given situation. So 617 00:33:13,200 --> 00:33:15,320 Speaker 1: it's very versatile in that way, which I thought was 618 00:33:15,400 --> 00:33:18,920 Speaker 1: really really cool. Now every single server in that peer 619 00:33:18,960 --> 00:33:22,040 Speaker 1: network can actually act as two different things. It can 620 00:33:22,080 --> 00:33:24,560 Speaker 1: act as an edge server or it can act as 621 00:33:24,560 --> 00:33:27,560 Speaker 1: an origin server. So edge servers these are the machines 622 00:33:27,600 --> 00:33:30,320 Speaker 1: that are actually serving up the video streams that we're watching. 623 00:33:30,560 --> 00:33:32,480 Speaker 1: So if you log into twitch tv so that you 624 00:33:32,480 --> 00:33:35,520 Speaker 1: can watch someone else, you're getting that video feed from 625 00:33:35,520 --> 00:33:38,239 Speaker 1: an edge server. Now, the person who is making that 626 00:33:38,400 --> 00:33:41,720 Speaker 1: video is sending it to what is called an origin server. 627 00:33:41,920 --> 00:33:45,080 Speaker 1: That's the that's the machine that's accepting the incoming video, 628 00:33:45,160 --> 00:33:47,400 Speaker 1: so it can distribute it to the edge servers. And 629 00:33:47,440 --> 00:33:50,280 Speaker 1: because all the machines can do either role, it makes 630 00:33:50,320 --> 00:33:53,600 Speaker 1: it an incredibly flexible approach to delivering live video. And 631 00:33:53,680 --> 00:33:56,840 Speaker 1: that's why twitch dot tv works. If it had a 632 00:33:56,920 --> 00:34:00,959 Speaker 1: designated origin server and then designated servers, you would be 633 00:34:01,000 --> 00:34:04,600 Speaker 1: limited by that capacity. Now, if you could actually visualize 634 00:34:04,640 --> 00:34:08,240 Speaker 1: all these connections it would look very much like a tree. 635 00:34:08,920 --> 00:34:12,719 Speaker 1: So as people start to tune into a particular video stream, 636 00:34:12,760 --> 00:34:15,239 Speaker 1: you would have more and more branches forming off of 637 00:34:15,280 --> 00:34:18,640 Speaker 1: the trunk of that tree, so that those servers would 638 00:34:18,640 --> 00:34:21,319 Speaker 1: be able to deliver up the video most effectively to 639 00:34:21,480 --> 00:34:25,080 Speaker 1: those users. Because of course users aren't centralized either, right, 640 00:34:25,360 --> 00:34:27,720 Speaker 1: You could have users from all over the world tuning 641 00:34:27,719 --> 00:34:30,520 Speaker 1: in for any given video stream. So you want to 642 00:34:30,520 --> 00:34:32,799 Speaker 1: make sure that you can deliver the video stream the 643 00:34:32,840 --> 00:34:36,200 Speaker 1: best possible way to any individual user, uh, in the 644 00:34:36,239 --> 00:34:38,719 Speaker 1: way that it's going to make financial sense. So I mean, 645 00:34:39,080 --> 00:34:41,880 Speaker 1: that's why Usher, I thought, is so impressive because it 646 00:34:41,920 --> 00:34:45,400 Speaker 1: can take all that into account. Now, how is this 647 00:34:45,560 --> 00:34:48,279 Speaker 1: actually working on the back end? Like what's going on? 648 00:34:48,440 --> 00:34:51,520 Speaker 1: So with YouTube, you know you're saving two disk, but 649 00:34:51,880 --> 00:34:55,719 Speaker 1: with Twitch you aren't Twitch and justin TV do not 650 00:34:56,000 --> 00:34:58,919 Speaker 1: save video to a disk to serve it up. They 651 00:34:59,000 --> 00:35:04,319 Speaker 1: save the video stream within the server's memory. So the 652 00:35:04,480 --> 00:35:07,080 Speaker 1: video that's going to the origin server is saved and 653 00:35:07,120 --> 00:35:09,760 Speaker 1: its memory, it is making copies of that video stream 654 00:35:09,880 --> 00:35:13,680 Speaker 1: as it's being generated, sends those copies to the edge servers. 655 00:35:13,920 --> 00:35:17,120 Speaker 1: Those copies stay in the edge server's memory, which then 656 00:35:17,200 --> 00:35:19,799 Speaker 1: make a copy of that stream and send it to 657 00:35:20,000 --> 00:35:23,160 Speaker 1: the clients, the browsers that people are using to watch 658 00:35:23,200 --> 00:35:25,480 Speaker 1: the video stream. So it's copies and copies and copies. 659 00:35:26,719 --> 00:35:29,239 Speaker 1: Now that's not to say that they don't save this 660 00:35:29,360 --> 00:35:32,400 Speaker 1: information to disk. They do, but that's for archival purposes, 661 00:35:32,440 --> 00:35:35,239 Speaker 1: so you can go back and watch a video that's 662 00:35:35,239 --> 00:35:39,240 Speaker 1: been recorded previously. But it's not that that disk saving 663 00:35:39,239 --> 00:35:42,440 Speaker 1: things is what's coming into play here exactly while you're 664 00:35:42,440 --> 00:35:44,759 Speaker 1: watching it live. It has nothing to do with the 665 00:35:44,800 --> 00:35:47,120 Speaker 1: fact that it's also being saved to disk. That's that's 666 00:35:47,120 --> 00:35:50,680 Speaker 1: a completely separate thing that's going on. So when you're 667 00:35:50,680 --> 00:35:55,279 Speaker 1: watching your browser, your your client is sending in a 668 00:35:55,320 --> 00:36:01,520 Speaker 1: request via HTTP, so hypertext protocol good, all fun stuff. 669 00:36:03,400 --> 00:36:06,200 Speaker 1: We'll do an episode on it someday, I promise. But anyway, 670 00:36:06,560 --> 00:36:10,040 Speaker 1: it sends the HTTP request and Usher then accepts the 671 00:36:10,080 --> 00:36:12,759 Speaker 1: request because you know, he's a good guy. Uh, then 672 00:36:12,840 --> 00:36:15,920 Speaker 1: he decided I'm sorry, I'm thinking the software Usher decides 673 00:36:15,960 --> 00:36:18,920 Speaker 1: which server to connect you to, which edge server you 674 00:36:18,960 --> 00:36:21,880 Speaker 1: would be connected to, and then the video servers themselves, 675 00:36:21,960 --> 00:36:24,799 Speaker 1: they're they're very simple. They're just either acting as edge 676 00:36:24,880 --> 00:36:27,560 Speaker 1: or origin servers, so they're not making this determination at all. 677 00:36:27,600 --> 00:36:31,760 Speaker 1: That's all the usher software stuff, so and multiple data centers. 678 00:36:32,280 --> 00:36:36,360 Speaker 1: The data centers are actually located in proximity to major 679 00:36:36,520 --> 00:36:40,799 Speaker 1: peering exchanges on the Internet. Now, appearing exchange is not 680 00:36:40,960 --> 00:36:44,719 Speaker 1: the same thing as peer to peer networks. It's two 681 00:36:44,719 --> 00:36:48,320 Speaker 1: different concepts. Appearing exchange is a point on the Internet 682 00:36:48,320 --> 00:36:52,680 Speaker 1: where two major networks end up connecting to each other. 683 00:36:52,719 --> 00:36:55,680 Speaker 1: So remember the Internet is a network of networks, right, 684 00:36:56,040 --> 00:36:58,880 Speaker 1: So when you've got two big networks, like two big 685 00:36:59,000 --> 00:37:02,800 Speaker 1: Internet service provid witers for example, to have a connection point, 686 00:37:03,400 --> 00:37:06,279 Speaker 1: the purpose for that is to allow stuff that came 687 00:37:06,360 --> 00:37:10,680 Speaker 1: from one network to cross over and enter another network. 688 00:37:10,960 --> 00:37:14,919 Speaker 1: And these period exchanges their agreements essentially that say, hey, 689 00:37:15,040 --> 00:37:17,279 Speaker 1: as long as you let the stuff that comes from 690 00:37:17,320 --> 00:37:20,560 Speaker 1: my network travel across your network, all of the stuff 691 00:37:20,600 --> 00:37:23,359 Speaker 1: that comes from your network travel across my network, and 692 00:37:23,360 --> 00:37:26,120 Speaker 1: we're all even, Stephen. Now, this is one of the 693 00:37:26,200 --> 00:37:30,200 Speaker 1: points that net neutrality really makes important. They say, this 694 00:37:30,280 --> 00:37:33,840 Speaker 1: relationship is why we don't have to worry about any 695 00:37:33,880 --> 00:37:36,840 Speaker 1: particular part of the network holding the rest of the 696 00:37:36,840 --> 00:37:40,160 Speaker 1: Internet hostage by saying, if you want your traffic to 697 00:37:40,200 --> 00:37:41,880 Speaker 1: travel on my network, you're gonna have to pay me 698 00:37:42,200 --> 00:37:46,319 Speaker 1: x amount of money per x amount of data. That 699 00:37:46,360 --> 00:37:49,520 Speaker 1: would be bad. That would that would I mean it 700 00:37:49,520 --> 00:37:51,279 Speaker 1: would be bad. It would break the Internet the way 701 00:37:51,280 --> 00:37:53,960 Speaker 1: we have it today because you would make an uneven 702 00:37:54,000 --> 00:37:56,719 Speaker 1: playing ground, and if other networks followed suit, it would 703 00:37:56,760 --> 00:38:00,080 Speaker 1: just become this thing where you would have your the 704 00:38:00,080 --> 00:38:02,440 Speaker 1: the stuff you would have access to would be completely 705 00:38:02,480 --> 00:38:07,839 Speaker 1: dependent upon what originated on your network, who how much, 706 00:38:08,040 --> 00:38:11,880 Speaker 1: or who wasn't paying who? Yeah. Yeah, So that's anyway 707 00:38:11,920 --> 00:38:15,839 Speaker 1: again tangential. But the reason why the peering exchanges are 708 00:38:15,880 --> 00:38:17,960 Speaker 1: important is that by putting your data centers close to 709 00:38:18,000 --> 00:38:22,719 Speaker 1: peering exchanges, you have a much closer connection to individual networks. 710 00:38:22,880 --> 00:38:25,480 Speaker 1: Because this is kind of like the crossroads. If you're 711 00:38:25,480 --> 00:38:28,000 Speaker 1: able to plant yourself near the crossroads, you can connect 712 00:38:28,000 --> 00:38:30,000 Speaker 1: to all those different roads that split off from it 713 00:38:30,200 --> 00:38:33,040 Speaker 1: very easily. It's a very strategic move on the part 714 00:38:33,040 --> 00:38:35,919 Speaker 1: of Justin dot tv and Twitch dot tv. So really 715 00:38:35,960 --> 00:38:39,120 Speaker 1: the only thing left to talk about is the user experience. 716 00:38:39,440 --> 00:38:43,800 Speaker 1: You've got a browser, You've got a embettable player inside 717 00:38:43,840 --> 00:38:46,759 Speaker 1: that browser. That's what's handling the incoming video to play 718 00:38:46,760 --> 00:38:49,480 Speaker 1: it on your your screen. Same sort of thing is 719 00:38:49,520 --> 00:38:52,160 Speaker 1: like an inbettable YouTube player or blit TV, I mean, 720 00:38:52,200 --> 00:38:55,600 Speaker 1: any of those inbeatable players are doing the basic same thing. 721 00:38:55,680 --> 00:38:58,279 Speaker 1: That's just that's just the interface that allows you to 722 00:38:58,360 --> 00:39:01,120 Speaker 1: watch whatever the data is. So really there's not a 723 00:39:01,160 --> 00:39:03,760 Speaker 1: whole lot going on the client side. There's some load 724 00:39:03,800 --> 00:39:06,000 Speaker 1: balancing stuff that comes from the client side, but not 725 00:39:06,040 --> 00:39:07,719 Speaker 1: by not a whole lot. Most of it's on the 726 00:39:07,760 --> 00:39:10,640 Speaker 1: back end. Um and and same thing for the broadcaster. 727 00:39:10,719 --> 00:39:12,800 Speaker 1: There's not a whole lot going on that the broadcaster 728 00:39:12,840 --> 00:39:15,000 Speaker 1: has to manage. They basically just turn on their webcam 729 00:39:15,360 --> 00:39:17,520 Speaker 1: and to use the client and go. Yeah, you might 730 00:39:17,560 --> 00:39:20,640 Speaker 1: have to have a little extra software depending upon where 731 00:39:20,719 --> 00:39:24,120 Speaker 1: your video is coming from, like UH. For example, when 732 00:39:24,120 --> 00:39:26,920 Speaker 1: I needed to stream when I was playing Minecraft on 733 00:39:26,960 --> 00:39:29,680 Speaker 1: the PC UH and I needed to stream that online, 734 00:39:29,719 --> 00:39:32,319 Speaker 1: I had to download another piece of software that would 735 00:39:32,360 --> 00:39:35,239 Speaker 1: allow me to take the video from the game and 736 00:39:35,360 --> 00:39:38,800 Speaker 1: upload it to twitch Live. But there are lots of 737 00:39:38,840 --> 00:39:40,759 Speaker 1: different ways of doing that. And then when you've got 738 00:39:40,760 --> 00:39:42,879 Speaker 1: things like the Xbox one and the PS four, it's 739 00:39:42,920 --> 00:39:46,600 Speaker 1: incorporated directly into the consoles through the firmware, so that 740 00:39:46,719 --> 00:39:49,319 Speaker 1: one that's taken care of it for you. Right. This 741 00:39:49,480 --> 00:39:51,680 Speaker 1: ease of usability, I think is why it is in 742 00:39:51,719 --> 00:39:54,200 Speaker 1: fact so stink and popular. Um and and you know 743 00:39:54,200 --> 00:39:57,000 Speaker 1: there's that forty five million unique users per month that 744 00:39:57,040 --> 00:40:00,239 Speaker 1: we mentioned earlier is really is nothing to sneeze. No, 745 00:40:00,360 --> 00:40:02,520 Speaker 1: that's huge. And and and it's not just that there are 746 00:40:02,520 --> 00:40:04,120 Speaker 1: a lot of them. It's like the other thing you 747 00:40:04,200 --> 00:40:06,400 Speaker 1: pointed out, Lauren, it's there's a lot of them that 748 00:40:06,440 --> 00:40:10,719 Speaker 1: are watching a really long time. Yeah. They boast that 749 00:40:10,920 --> 00:40:13,640 Speaker 1: per week as a fall that was more viewers than 750 00:40:13,719 --> 00:40:17,560 Speaker 1: Breaking Bed, Tosh point O and the NFL preseason combined. 751 00:40:17,760 --> 00:40:21,239 Speaker 1: That's yeah, I mean that's a significant audience. And you know, 752 00:40:21,320 --> 00:40:24,719 Speaker 1: again they're really engaged. I mean, if I go to 753 00:40:24,800 --> 00:40:26,880 Speaker 1: YouTube and I'll watch a video, a lot of those 754 00:40:26,960 --> 00:40:29,160 Speaker 1: videos tend to be three to ten minutes long. You know, 755 00:40:29,239 --> 00:40:31,200 Speaker 1: there's some that are longer, and I'll watch those two 756 00:40:31,239 --> 00:40:34,239 Speaker 1: depending upon what I'm interested in. You go, maybe a 757 00:40:34,280 --> 00:40:36,040 Speaker 1: link along the side catches your eye and you watch 758 00:40:36,080 --> 00:40:38,279 Speaker 1: something else, but probably you're going to click away. Yeah, 759 00:40:38,320 --> 00:40:40,879 Speaker 1: you don't necessarily stay there for you know, more than 760 00:40:41,280 --> 00:40:43,520 Speaker 1: one or two videos. If you do, I mean, that's 761 00:40:43,520 --> 00:40:47,320 Speaker 1: cool too, but that's probably not the average person's experience. 762 00:40:47,680 --> 00:40:49,359 Speaker 1: You know, like a lot of people watch a YouTube 763 00:40:49,440 --> 00:40:52,560 Speaker 1: video when it's embedded in something else like a Facebook post. 764 00:40:52,719 --> 00:40:55,040 Speaker 1: The typical Twitch user, on the other hand, will stay 765 00:40:55,080 --> 00:40:57,239 Speaker 1: on the site for hours that that average of a 766 00:40:57,320 --> 00:41:00,280 Speaker 1: hundred and six minutes per day that we mentioned are earlier. 767 00:41:00,960 --> 00:41:02,719 Speaker 1: If that's an average, that means there are a lot 768 00:41:02,800 --> 00:41:06,640 Speaker 1: of people watching way longer than that. Oh right, yeah. Um, 769 00:41:06,800 --> 00:41:09,200 Speaker 1: And and the demographic that's going to twitch is a 770 00:41:09,239 --> 00:41:12,160 Speaker 1: really desirable one. The majority is under thirty five. I 771 00:41:12,160 --> 00:41:14,800 Speaker 1: think it's twenty one to twenty nine is the main, 772 00:41:15,120 --> 00:41:19,200 Speaker 1: main portion that that's the group of people that you say, hey, 773 00:41:19,239 --> 00:41:22,239 Speaker 1: these people buy stuff. Yeah, you've got lots of money 774 00:41:22,320 --> 00:41:24,959 Speaker 1: and you're probably willing to give it to me. Yeah, 775 00:41:25,000 --> 00:41:27,279 Speaker 1: so you aren't smart enough yet to know not to know. 776 00:41:27,400 --> 00:41:30,879 Speaker 1: I'm just kidding. I'm just kidding. But yeah. And it's 777 00:41:30,920 --> 00:41:35,239 Speaker 1: also largely male, although there's a growing audience and broadcast 778 00:41:35,280 --> 00:41:38,440 Speaker 1: group of ladies as well. So we're starting to see 779 00:41:38,719 --> 00:41:42,840 Speaker 1: that that demographic is slowly changing. I mean, obviously gaming 780 00:41:42,880 --> 00:41:47,280 Speaker 1: has been very much male domin Yeah, but we're starting 781 00:41:47,320 --> 00:41:50,000 Speaker 1: to see a lot of females get into it because 782 00:41:50,360 --> 00:41:53,840 Speaker 1: they find gaming just as awesome of course, And and 783 00:41:53,920 --> 00:41:57,600 Speaker 1: as as the community changes and as the general standards 784 00:41:57,600 --> 00:42:01,160 Speaker 1: of the industry change, all of that is hopefully fingers 785 00:42:01,200 --> 00:42:05,399 Speaker 1: crossed going to change gender lines, right, So it's it's 786 00:42:05,440 --> 00:42:08,560 Speaker 1: again a growing market, So that's another reason why it's desirable. 787 00:42:09,000 --> 00:42:11,040 Speaker 1: But I mean that kind of engagement where you're going 788 00:42:11,080 --> 00:42:14,520 Speaker 1: to stay focused on something for on average a hundred 789 00:42:14,600 --> 00:42:19,280 Speaker 1: and six minutes. That is why Google is so interested 790 00:42:19,520 --> 00:42:22,400 Speaker 1: in in twitch dot tv. Not only is it lives, 791 00:42:22,440 --> 00:42:24,719 Speaker 1: not only have they done live streaming, and they've done 792 00:42:24,719 --> 00:42:29,439 Speaker 1: it well, they keep people there. Yeah. Yeah, as of now, 793 00:42:29,560 --> 00:42:33,000 Speaker 1: YouTube has really only been toying with live streaming. It 794 00:42:33,239 --> 00:42:36,960 Speaker 1: just opened their YouTube Live services up to the general 795 00:42:37,040 --> 00:42:41,319 Speaker 1: population in December and they haven't really done I mean, 796 00:42:41,320 --> 00:42:43,279 Speaker 1: it hasn't really seen that much traffic. Yeah. The only 797 00:42:43,280 --> 00:42:46,440 Speaker 1: other thing you can use is like the Google Hangouts, 798 00:42:46,520 --> 00:42:49,839 Speaker 1: where you can do a hangout broadcast and that ends 799 00:42:49,920 --> 00:42:53,320 Speaker 1: up broadcasting on YouTube as well, plus on Google Plus 800 00:42:53,320 --> 00:42:55,320 Speaker 1: and that kind of stuff. Um, and I've used that 801 00:42:55,360 --> 00:42:56,840 Speaker 1: a couple of times when I've been a guest on 802 00:42:56,920 --> 00:43:00,120 Speaker 1: other shows. Really I wasn't the one using it, they was. 803 00:43:00,239 --> 00:43:03,640 Speaker 1: I was participating in it. But yeah, it's still like 804 00:43:03,640 --> 00:43:06,040 Speaker 1: you said, it's a niche compared to someone like Twitch 805 00:43:06,080 --> 00:43:09,359 Speaker 1: dot tv that has just this huge audience. So what 806 00:43:09,360 --> 00:43:14,600 Speaker 1: would it mean if Google and Twitch joint forces to 807 00:43:14,760 --> 00:43:17,200 Speaker 1: rule the galaxy. Well, first of all, they might not 808 00:43:17,320 --> 00:43:20,200 Speaker 1: be able to. On a federal level, The Feds might 809 00:43:20,239 --> 00:43:25,040 Speaker 1: try to block this deal based on anti competition by saying, hey, Google, 810 00:43:25,080 --> 00:43:27,880 Speaker 1: you've already got YouTube, You've got you've got user generated 811 00:43:28,040 --> 00:43:31,040 Speaker 1: video sewn up pretty much. Yeah, you're the biggest content 812 00:43:31,160 --> 00:43:33,560 Speaker 1: player and you're trying to merge with the biggest content 813 00:43:33,640 --> 00:43:36,560 Speaker 1: streamer and that's really not okay. So that's a pity. 814 00:43:36,560 --> 00:43:38,920 Speaker 1: So that's a possibility. But if it goes through, it 815 00:43:38,960 --> 00:43:43,520 Speaker 1: would mean access to Twitch users to Google's incredible infrastructure, 816 00:43:43,920 --> 00:43:47,759 Speaker 1: um therefore improving the video quality and uh and lag 817 00:43:47,840 --> 00:43:51,200 Speaker 1: times that users are currently experiencing, which I mean Twitch 818 00:43:51,239 --> 00:43:54,120 Speaker 1: has said themselves that they were not prepared for the 819 00:43:54,200 --> 00:43:56,000 Speaker 1: kind of growth that they have seen and that it's 820 00:43:56,000 --> 00:43:59,640 Speaker 1: a fabulous problem to have. But yeah, it's still a problem. 821 00:43:59,640 --> 00:44:02,400 Speaker 1: It's still problem. So, like I said, Google has a 822 00:44:02,400 --> 00:44:05,640 Speaker 1: lot of experience with this kind of peering relationship, using 823 00:44:05,640 --> 00:44:08,440 Speaker 1: it for other things as well. So h So, in 824 00:44:08,800 --> 00:44:13,400 Speaker 1: at least on a philosophical level, it's from from architecture, 825 00:44:13,560 --> 00:44:18,120 Speaker 1: it's a good fit. I don't know how close twitches 826 00:44:18,239 --> 00:44:21,040 Speaker 1: architecture resembles that of Google on a practical level, but 827 00:44:21,080 --> 00:44:24,040 Speaker 1: philosophically they are very similar, right, and it's so it 828 00:44:24,080 --> 00:44:27,560 Speaker 1: could mean really good things there. It could hypothetically mean 829 00:44:27,719 --> 00:44:30,880 Speaker 1: more money for the people who are broadcasting. Twitch currently 830 00:44:30,920 --> 00:44:35,680 Speaker 1: has some four thousand partner casters who who are paid 831 00:44:35,880 --> 00:44:37,960 Speaker 1: for their casts. Yeah, when you get a certain number 832 00:44:38,120 --> 00:44:41,400 Speaker 1: of subscribers, you tend to get the attention of Twitch. 833 00:44:41,880 --> 00:44:44,880 Speaker 1: So let's say that you suddenly hits some new thousand subscribers, 834 00:44:44,880 --> 00:44:48,880 Speaker 1: which is going to say, hey, we should work together. Yeah, 835 00:44:48,920 --> 00:44:51,960 Speaker 1: this could be beneficial for everyone. Um, but Google could 836 00:44:52,040 --> 00:44:55,799 Speaker 1: possibly do even more for these content creators and if 837 00:44:55,800 --> 00:44:59,960 Speaker 1: nothing else, integrate more seamlessly with searching with YouTube, increasing 838 00:45:00,320 --> 00:45:06,759 Speaker 1: both ad revenue and audience participation. Alternately, it could mean 839 00:45:06,960 --> 00:45:10,840 Speaker 1: changes to the contracts of these professional live streamers, especially 840 00:45:10,840 --> 00:45:15,520 Speaker 1: considering YouTube's pretty strict policies about copyrighted material and on 841 00:45:15,840 --> 00:45:19,040 Speaker 1: region locking. Yeah. I mean a thing that a lot 842 00:45:19,120 --> 00:45:22,680 Speaker 1: of streamers do is they'll they'll have live music. Well 843 00:45:22,680 --> 00:45:24,160 Speaker 1: not live music. I don't think that they have a 844 00:45:24,160 --> 00:45:26,280 Speaker 1: band in their living right, but they have sometimes playing 845 00:45:26,280 --> 00:45:28,880 Speaker 1: in the background while they are playing a game, and 846 00:45:28,960 --> 00:45:31,120 Speaker 1: YouTube is not excited about that sort of thing. Well, 847 00:45:31,160 --> 00:45:33,279 Speaker 1: more importantly, the music in this reason not I'm not 848 00:45:33,320 --> 00:45:36,359 Speaker 1: excited about that. They don't necessarily want their music associated 849 00:45:36,360 --> 00:45:41,399 Speaker 1: with There's also some control issues, like Nintendo historically has 850 00:45:41,400 --> 00:45:45,520 Speaker 1: been very very protective of their of their video games, 851 00:45:45,560 --> 00:45:49,160 Speaker 1: and people who have put up, uh YouTube videos of 852 00:45:49,200 --> 00:45:51,560 Speaker 1: them playing the Nintendo games, a lot of them have 853 00:45:51,719 --> 00:45:56,399 Speaker 1: experienced takedown because Nintendo said, you know what, we want 854 00:45:56,440 --> 00:45:59,040 Speaker 1: to have full control of this experience. We don't like 855 00:45:59,200 --> 00:46:02,760 Speaker 1: the idea of other people being able to do that. 856 00:46:02,880 --> 00:46:05,239 Speaker 1: Nintendo started to ease up a little bit on that 857 00:46:05,440 --> 00:46:08,799 Speaker 1: more recently, but it's traditionally been an issue, and so 858 00:46:08,880 --> 00:46:12,600 Speaker 1: there's no guarantee that every single game publisher is going 859 00:46:12,640 --> 00:46:15,760 Speaker 1: to be totally cool. That's a whole that's a whole 860 00:46:15,840 --> 00:46:18,560 Speaker 1: other issue that I really wish that Nintendo would get with. 861 00:46:19,360 --> 00:46:21,880 Speaker 1: But back back on on the Google deal side, it 862 00:46:22,200 --> 00:46:26,359 Speaker 1: could hypothetically get other demographics than just gamers into this 863 00:46:26,480 --> 00:46:29,680 Speaker 1: live internet streaming kind of stick. Yeah, it's kind of 864 00:46:29,719 --> 00:46:33,240 Speaker 1: the whole purpose of justin dot tv in the first place. 865 00:46:33,680 --> 00:46:35,719 Speaker 1: But Twitch dot tv was the one that took that 866 00:46:35,800 --> 00:46:37,799 Speaker 1: was the part that took off off of justin dot tv. 867 00:46:37,880 --> 00:46:41,640 Speaker 1: And now now that people are getting more familiar or 868 00:46:41,719 --> 00:46:44,720 Speaker 1: aware of this, and also now that Twitch has grown 869 00:46:44,760 --> 00:46:48,480 Speaker 1: it into a very usable process, yeah, that we might 870 00:46:48,520 --> 00:46:53,520 Speaker 1: see more people get into this kind of live broadcast platform. Yeah, 871 00:46:53,560 --> 00:46:56,600 Speaker 1: for for events large and small, for for entertainment, maybe 872 00:46:56,600 --> 00:47:01,359 Speaker 1: for for science kind of discovery broadcasts, for news. I've 873 00:47:01,400 --> 00:47:04,239 Speaker 1: seen a couple of podcasts use this sort of thing 874 00:47:04,360 --> 00:47:07,000 Speaker 1: as to live stream when they were actually recording, so 875 00:47:07,080 --> 00:47:10,280 Speaker 1: you could follow along. Now, before any of you ask, 876 00:47:10,800 --> 00:47:13,839 Speaker 1: that's not gonna happen with us. I think in this 877 00:47:13,920 --> 00:47:17,440 Speaker 1: in this recording alone, we've stopped to cough about nineteen times. 878 00:47:17,480 --> 00:47:19,480 Speaker 1: I in fact, had to stop to cough before I 879 00:47:19,520 --> 00:47:25,319 Speaker 1: recorded this line. She actually started the line, stopped coughed. Yeah, no, 880 00:47:25,320 --> 00:47:29,080 Speaker 1: no one, no one wants that we both have congestion issues. 881 00:47:29,200 --> 00:47:31,399 Speaker 1: So yeah, if you guys were watching live right now, 882 00:47:31,760 --> 00:47:35,160 Speaker 1: it would be cough cough cough cough cough, three paragraphs 883 00:47:35,200 --> 00:47:38,279 Speaker 1: worth of information cough cough cough. Yeah, that that how 884 00:47:38,320 --> 00:47:42,120 Speaker 1: it works. Section took us forever because of me, So yeah, 885 00:47:42,280 --> 00:47:44,120 Speaker 1: we will not be doing that. And even when we're 886 00:47:44,120 --> 00:47:47,920 Speaker 1: feeling well, still not doing it so, but but at 887 00:47:47,920 --> 00:47:49,920 Speaker 1: any rate, I mean, you know this, this could this 888 00:47:49,920 --> 00:47:53,520 Speaker 1: could hypothetically change television as we know it. This could 889 00:47:53,560 --> 00:47:57,760 Speaker 1: be the bridge between network broadcasts and internet viewing. Yeah, 890 00:47:57,880 --> 00:48:01,720 Speaker 1: we could actually see very re regionalized approaches to stuff 891 00:48:01,760 --> 00:48:05,680 Speaker 1: as well as you know, more global appeal kind of broadcasts. 892 00:48:06,520 --> 00:48:10,800 Speaker 1: I've seen some communities do kind of a local news 893 00:48:10,840 --> 00:48:13,760 Speaker 1: broadcast for that community. This sort of thing would allow. 894 00:48:13,840 --> 00:48:16,720 Speaker 1: It's kind of like the public access in that sense, 895 00:48:16,719 --> 00:48:20,719 Speaker 1: and that it's a live way to communicate information out 896 00:48:20,760 --> 00:48:25,120 Speaker 1: whether it's entertainment or educational or news or whatever. You 897 00:48:25,160 --> 00:48:28,879 Speaker 1: can do that on a specific schedule. The other advantageos, 898 00:48:28,920 --> 00:48:31,880 Speaker 1: of course, it does record the videos, so even if 899 00:48:31,880 --> 00:48:33,680 Speaker 1: you aren't able to tune in live, you can still 900 00:48:33,719 --> 00:48:36,520 Speaker 1: watch it, which gives an advantage over other types of 901 00:48:36,560 --> 00:48:39,759 Speaker 1: live broadcasts. Sure, and it's it's on demand all the time. 902 00:48:39,760 --> 00:48:42,839 Speaker 1: It's on the internet. Anyone in the world hypothetically can 903 00:48:42,920 --> 00:48:45,480 Speaker 1: tune into this. Yeah. So let's say that you move 904 00:48:45,560 --> 00:48:47,720 Speaker 1: away from your hometown, but you want to keep tabs 905 00:48:47,800 --> 00:48:50,480 Speaker 1: on it. That could potentially be a way of that happening. 906 00:48:50,880 --> 00:48:54,440 Speaker 1: Or if one performer becomes very popular and and explodes 907 00:48:54,520 --> 00:48:57,719 Speaker 1: across you know, into other communities. That's wonderful. We're still 908 00:48:57,760 --> 00:49:03,240 Speaker 1: waiting make us explode people, you know, in popularity that is, yes, please, 909 00:49:03,239 --> 00:49:06,279 Speaker 1: thank you. Most of my atoms where they are. Yeah, 910 00:49:06,400 --> 00:49:10,120 Speaker 1: I'm pretty fond of where my my constituent atoms happen 911 00:49:10,160 --> 00:49:13,040 Speaker 1: to be as well. So anyway, that's that's the lowdown 912 00:49:13,160 --> 00:49:16,320 Speaker 1: on Twitch dot tv. It'll be really interesting to see 913 00:49:16,440 --> 00:49:21,720 Speaker 1: how this Xbox and PlayStation incorporation ends up, um shaking out. Yeah, 914 00:49:21,760 --> 00:49:23,680 Speaker 1: and whether or not the Google deal goes through. Of course, 915 00:49:23,719 --> 00:49:26,040 Speaker 1: all of you guys know possibly by the time this 916 00:49:26,080 --> 00:49:29,200 Speaker 1: podcast goes live, whether or not that happened, we are 917 00:49:29,200 --> 00:49:31,839 Speaker 1: still in the dark as of the recording, so I'm 918 00:49:31,880 --> 00:49:34,759 Speaker 1: really curious to see how it works out. Yeah. I 919 00:49:34,800 --> 00:49:37,400 Speaker 1: really think it's an exciting development. Thank you so much 920 00:49:37,480 --> 00:49:40,279 Speaker 1: Daniel for the question. Yeah. And and also I mean, 921 00:49:40,320 --> 00:49:42,640 Speaker 1: you know, for for everyone out there who uses Twitch 922 00:49:42,760 --> 00:49:46,759 Speaker 1: and is worried about the deal, uh, you know, I 923 00:49:47,200 --> 00:49:50,600 Speaker 1: understand it's it's scary whenever Google buys a property that 924 00:49:50,640 --> 00:49:53,880 Speaker 1: you are very fond of, especially giving Google's habit of 925 00:49:54,320 --> 00:49:56,200 Speaker 1: you know, saying huh, I don't know how to do 926 00:49:56,239 --> 00:49:59,560 Speaker 1: this anyway, and then shut it down. Yeah. Yeah, but 927 00:49:59,680 --> 00:50:02,040 Speaker 1: you know, I think that it could very much be 928 00:50:02,160 --> 00:50:05,920 Speaker 1: an excellent opportunity for many industries. So I think. Uh. 929 00:50:06,280 --> 00:50:08,239 Speaker 1: The other important thing to take away from this is 930 00:50:08,280 --> 00:50:11,000 Speaker 1: the idea of Microsoft trying to buy Twitch and being 931 00:50:11,000 --> 00:50:14,200 Speaker 1: turned down should give all of you Sony PlayStation four 932 00:50:14,239 --> 00:50:17,319 Speaker 1: owners a big sigh of relief. I mean, I can 933 00:50:17,400 --> 00:50:20,480 Speaker 1: only imagine if and Google has someone of a vested 934 00:50:20,520 --> 00:50:23,520 Speaker 1: interest in this. But Google doesn't have its own gaming 935 00:50:23,600 --> 00:50:27,400 Speaker 1: console out. You could argue that Android is a platform 936 00:50:27,480 --> 00:50:30,080 Speaker 1: for games, but it's just one platform. It's not the 937 00:50:30,480 --> 00:50:33,719 Speaker 1: main purpose for Android. Um. But you know, since they 938 00:50:33,760 --> 00:50:37,440 Speaker 1: don't have a vested interest necessarily, and whether Microsoft succeeds 939 00:50:37,480 --> 00:50:40,600 Speaker 1: over Sony, you don't have to worry about Twitch suddenly 940 00:50:40,640 --> 00:50:43,200 Speaker 1: losing support for something like the PlayStation four, which is 941 00:50:43,239 --> 00:50:46,440 Speaker 1: something I would legitimately worry about if Microsoft had bought Twitch, 942 00:50:46,719 --> 00:50:48,800 Speaker 1: I would really be curious to see how they would 943 00:50:49,200 --> 00:50:53,279 Speaker 1: resist that urge. So at least that that danger is 944 00:50:53,280 --> 00:50:56,400 Speaker 1: not on the table anymore. Daniel, thank you again for 945 00:50:56,480 --> 00:50:59,080 Speaker 1: your suggestion. If you would like to send us a 946 00:50:59,080 --> 00:51:02,040 Speaker 1: message like Daniel did, well, if you want to be 947 00:51:02,080 --> 00:51:04,359 Speaker 1: just like Daniel, you need to do the email thing 948 00:51:04,640 --> 00:51:08,719 Speaker 1: and our email addresses tech stuff at Discovery dot com, 949 00:51:08,840 --> 00:51:11,200 Speaker 1: or you can drop us a line on Facebook or Twitter. 950 00:51:11,280 --> 00:51:13,560 Speaker 1: Our handle is tech stuff hs W. Hey, we got 951 00:51:13,600 --> 00:51:16,719 Speaker 1: a tumbler too, which is awesome. Text Stuff HSW for 952 00:51:16,760 --> 00:51:19,400 Speaker 1: that as well, so check that out. Leave us a message, 953 00:51:19,480 --> 00:51:21,960 Speaker 1: let's know what you think, give us suggestions for topics, 954 00:51:22,000 --> 00:51:27,400 Speaker 1: and we will talk to you again really soon for 955 00:51:27,520 --> 00:51:29,840 Speaker 1: more on this and thousands of other topics because it 956 00:51:29,920 --> 00:51:40,880 Speaker 1: how staff works dot com