Google: Let’s Talk About Web 3.0

Gary 0:10
Welcome, everyone to the next episode of the search off the record podcast. Our plan is to talk a bit about what’s happening at Google search, how things work behind the scenes, and maybe just maybe have some fun along the way. My name is Gary Illyes. And I am a search advocate on the search relations team here at Google in Switzerland. I’m joined today by Martin and Alan, both also on the search relations team. Alan’s focus on our team is e-commerce in Google Search, say hi, people. Hi. That was Martin. Good day. That was not Martin, that was Alan. Okay. Today, we thought we could talk about something exciting that’s happening on the internet. And that is web three. Today’s episode is a little bit different, because we are not going to talk about an official position of Google on web three. But rather, it’s our thoughts about web three. And this whole topic started from discussion that we started in a team meeting a few weeks ago. And we just stopped the discussion and decided to do it in public. Well, still off the record. But in the public, nonetheless.

Martin 01:26
What could possibly go wrong?

Gary 01:29
Martin, please don’t just don’t. So before we jump into web three, maybe we should define the different stages of the web? Because it is kind of confusing sometimes. And maybe some people don’t know that there are different pseudo versions of the web. So maybe let’s talk about web one. What do you think web one means?

Alan Kent 1:54
My first recollection of web one was actually back in my university days, and the internet was just starting out, who am I dating how old I am, we won’t worry about that detail. And it was sort of like, oh, I want to find a manual for the this new bit of software. And oh, you have all these shelves of printed documentation. And it took forever to get them up to date. And web one to me was really about publishers making this content available on the web, you can get an update to a manual, you could search and find the manual, even if you didn’t have the printed version. And that was a big step forward at the time. But the content, it was all static. People could read it, but you couldn’t do anything with it. Other than read it, you couldn’t comment on it. It was just a publisher model. And some people you referred to that then as sort of the era of the read only web, people sharing content online. And other people could consume it. It was great, save trees, less less print outs, less these thousand-page manuals getting printed all over the shelves. And from that point of view, I think it was a great step forward. At least for web one.

Martin 2:55
Yeah. And I mean, back in the days, it was also costly. It was simply costly that the technology was relatively affordable and simple. But unless you had someone with the means of storing your stuff online somewhere, like university or something, it was barely affordable, like you would have to put money in to run a website, and significant money at that. Domains were significant, I think the price as well. And yeah.

Gary 3:21
Do you remember what was your first encounter with web one? Air quotes?

Martin 3:25
I think that was when I started working part time after school for a company in town that had a website. That was wild.

Gary 3:35
But there was no interaction with the website is basically it was just you go into the website, and then just reading and that’s it and leaving.

Martin 3:43
Yeah, yeah. Or from from my perspective, as a webmaster, as we used to be called, you would put stuff on the website. And then people could consume that content, roughly like broadcast media works as well, like, you can’t really you don’t have a back channel.

Gary 3:59
And then in comparison, web two would be more read and write in the sense that you as a user could leave a comment or you could even create your own content, instead of having someone create the content and then you just consume it.

Alan Kent 4:16
Yeah, I don’t remember checking the dates. But it feels like the era of the rise of WordPress and other similar CMS is where people could see comments and likes, that was an obvious first one, but then be able to write your own blog. And you can just do it straight through a web browser. That was pretty radical at the time,

Martin 4:34
and also a bunch of hosted services. So you didn’t–if you wanted to have a blog at the beginning, you would have to actually have a web server or web space. And then if you wanted one with a database that cost you extra and yada yada yada. You might run your blog from I don’t know Front Page or Dreamweaver or something. But that was tedious because then you would have to write HTML, but then then came I think, like life journal and Google blogs and all these other services where I just sign up with my email address or password or username, and then I get username, dot whatever service.com. And then I can just write stuff in the kind of What You See Is What You Get fashion.

Alan Kent 5:12
Yeah. And then I think the next generation or started to become the social platforms, I can publish something, and I don’t have to have people find me because I can be on the social platform, I can build a relationship between friends, and so they can know about it, and they can start sharing it. And the difference there, I think, is you got a built in audience and distribution mechanism. And frequently, the platforms would solve a monetization problem. Like if you’re trying to get money from your content, platforms worry about showing ads, and so forth. And they’d give a cut of that to the creators of the content.

Gary 5:47
I’m guessing that these platforms are also or were also, I guess, they still are good for, for example, small businesses, who don’t want to deal with setting up a server or figuring out how to deploy a WordPress installation. And they can just go to a platform like wordpress.com and sign up with a free account. And then they have their website with one click essentially, and then they just have to put up the content. So from from that perspective, it feels like these platforms solved a problem for a large portion of the web, or potential web, where it enabled people to create content, even if they didn’t have the background for creating websites, or the

Martin 6:29
Or the resources.

Gary 6:31
Or the resources, yes, because now it became a lot more accessible.

Alan Kent 6:33
Yeah. If you think of media like video, I mean, it’s not easy to host video and do a good job. And there’s lots of bandwidth issue and lots of storage issues. So

Gary 6:43
it’s actually a frickin nightmare. Yeah, like to host video, because like hosting images is kind of easy, because you have, let’s say, like, up to 1.5 megabyte image, let’s say that’s your limit on the image hoster. But with videos, you have to keep connections open for a very long time, usually, unless you split it, or you have to multiplex the video, for whatever reason. It’s actually very, very complicated.

Alan Kent 7:09
I think also, and it’s possibly towards, I don’t know if it can call it towards the end. But the other aspect of the monetization sort of stuff is people started to do things like, Well, how do I get paid other than just doing ads, and you started getting into Merchandising, and tipping and Patreon sites and Kickstarter projects, and so forth. To me, it was always been interesting, comparing also to China, because China really sort of came late to the game, I think, for the internet for the mass, per thing. But it started on mobile. And so getting ads on your page is a much worse experience on a mobile device. And so they there’s actually a lot of interesting innovation, it’s actually come out of China. And the whole area of monetization, if you look into it, because they couldn’t rely on ads as being the primary source of revenue for these platforms.

Gary 7:59
If you ever use the web on China, like WeChat, or anything like that, it’s actually fascinating because like, quite literally with a tap on the screen, you can pay for a service without actually leaving any window. It’s it’s so seamless, when you’re interacting, the I’m air quoting here, local internet, I actually love that part of it. And I hope that we can figure out how to do that on on a larger scale, not just in, in just one country. When you mentioned tipping, I was thinking of Twitch, which is a platform where people can stream video games, or woodworking or cooking or whatever. And every now and then it’s, it feels so nice to just go and find creators who don’t have viewers, and then just leave a tip for them like, I don’t know, like $1 or $2, or $3. And some of them are so excited about that. Because it’s probably also very hard to get followers on these platforms. Very often it might feel that you are talking to avoid and well, it’s just hard to get the followers and the viewers but then you have the platforms that actually help or may be able to help with that. And digg.com, d, i, g, g, .com back in the days was one of those where you could get easily hundreds or thousands of people onto your blog just by submitting your interesting air quoting again, content to that sharing site or search engines if you know any of them like a good one. They can send you traffic every now and then.

Alan Kent 9:29
Yeah, if only we knew someone who streamed on Twitch.

Martin 9:33
I really really tried not to say anything here but yeah, I do remember how great it felt when I saw I do stream live coding on open source every now and then. And it felt so nice to see like people popping up and it has been mostly people from my network more or less I did put it out on Twitter. When I did that and then people from who follow me on Twitter already followed me on on Twitch as well, it feels it feels nice. And it’s it’s bizarre because it’s this enclosed space and the web is so vast, and it’s so hard to get a little bit of like fellowship or like people seeing you on the on the general web on these platforms, it can be a little easier, because they are made to promote connections, and they’re made to promote discovery. So it feels nice. It’s always feels nice when there’s Twitch followers, or like someone popping into the Twitch stream, who I don’t know before. That’s, that’s lovely.

Gary 10:34
So we defined web one, we pretty much defined web two as well. Which means that we can probably move on to web three and try to define it. And I’m emphasizing try here. Because before this episode, I was doing some research, basically, just to not sound too damp on the podcast. And it feels to me that web three is not too well defined. Am I right on that?

Martin 11:02
Let’s be honest, web two itself is also not very well defined. Because depending on who you ask about Web 2.0, it meant different things. Like we said, the read write read, but then a lot of people say no, it’s only the social web. Other people say no, it’s the web of services like YouTube, Facebook, MySpace, whatnot. So I guess web three will be as fuzzy, now? Well, I

Alan Kent 11:29
certainly think some people are trying to define the web. What web three, zero means. And if, for example, I take the crypto space, it’s a lot of people come in about crypto and sort of say, hey, is this more of a marketing push as distinct from a really true name? Because like the names web one and web two more came about because you look at history, like you look back and you sort of say I can see a pattern there. We’ll call that web one. And I can see a new pattern web two. It didn’t define let’s do web two. And suddenly the web became that, no, it was more sort of saying this is what the web has become. And here’s a way of just talking about describing it. So but certainly crypto is one of those areas that has been pitched as web 3.0. And it’s sort of it comes about from I think, partly from the decentralized web definition and saying, Hey, crypto payments, they’re a distributed decentralized payment. So shouldn’t that be web three? I don’t fall into that camp myself. And there’s a lot of voices who don’t like that definition. Yeah, me. Yeah. Okay, we got some on this call too. Same.

Martin 12:36
So so here’s my problem with it. So and I came across web three, the first time, long after I came across people who work on decentralized web technologies. And that’s, that’s what what freaks me out about it is that if you look at, for instance, the Wikipedia definition, which starts off right off the bat with like, web based, web, the web based on blockchain technology, incorporating decentralization and token based economy, so basically, crypto and blockchain. And I don’t know, but like 90% of the cases where someone yells blockchain is the solution. It’s not. And crypto has its own challenges. And then it starts off with so what, let’s put aside the fact that there are, and have been decentralized efforts long before web three was coined. And they still go on and they still exist. And I still think they’re amazing one to mention, for my perspective, would be the interplanetary file system, or IPFS. For short. That’s cool stuff. That’s really really cool stuff based on Mesh routing and distributed hash tables. They don’t need a blockchain because by definition, their stuff is distributed. And I don’t know if they are now also using a blockchain because I haven’t followed their efforts in a while. But the blockchain is fantastic, where you don’t want a central authority to agree on things. And I don’t think for for most of the web things you’d need an authority to begin with. And you don’t need to agree on things like I want to pull like this collaborative players. By definition, I want to pull information from your site, you want to provide me this information. Now I see that we do want the blockchain for the financial side of things, maybe, okay, sure. But that’s a completely different title. Like for me, web monetization is one topic. And then decentralization is another thing. And then the question that I asked myself, but and I think you brought this up earlier as well. Before we started recording, is isn’t the web decentralized? What what do you think is the web by default decentralized?

Alan Kent 14:50
Well, from my point of view, there’s different definitions of what decentralized really means, like, from day one, there were web servers all around the world, and that’s what made the web special and Google search would come along and would help people find them. And that’s what search was about. Search didn’t host the content, it directed you people to the content out there on the web. And that was a useful service. And and all the search engines, of course. To me, I think what people mean, when I talk about decentralized web, it’s not that the services are distributed around the world, because they always have been. I think it’s more about the idea of communities. And you start getting into, I think about my kids you know, they’re on Discord servers around the place. And they hang out there with a group of friends. And they talk about one topic, they might be talking about the games they’re playing together, it might be a group of school kids, and so forth. But it’s really their community is the aspect of the decentralized because the thing about Discord in particular, for example, is it’s not public. It’s not on the open web, you can’t search and find your content on it. And so you get these small communities. And so to me, decentralized is more about, I’ve got this content, and only members of that community have got access to that content. And so it’s a more personal relationship with it. And you’ve got many of these sorts of communities. And to me, that’s really what decentralized is more about. But…

Gary 16:16
…isn’t that the textbook definition of walled gardens?

Alan Kent 16:19
It’s an interesting question. And walled gardens usually to me, incurs the idea of paywall, and these communities are more a matter of, I’m going to have this group of people that I trust that I can share my opinions. And because it’s a community in a smaller group, I’m going to be more open with them. And so to me, that’s the more the intent, or the feeling of the decentralized community is more, it’s the community aspect. It’s not that it’s a paywall in front of it. Now, you can disagree with that sort of thing. But that’s sort of the feeling I get between them.

Martin 16:55
The awkward feeling I have about these, and I will name them walled gardens, because to me, they are walled gardens, and it doesn’t matter who runs them. It can be YouTube, YouTube, is a walled garden. If I upload my content to YouTube, it is in YouTube’s hands. And that’s us, Google. And I don’t have a problem with Google doing that, because they provide me a lot of, as Gary said, video hosting is a pain in the lower back. And they take that away from me. So I can create content on YouTube, the problem there is that everything that I create, my community that I create, and plant into a walled garden doesn’t matter, which can be Facebook can be MySpace, doesn’t matter is in the house on the property of someone that can make whatever rules they want. And they can change these rules. So if I take my community, they’re trusting my host to be playing nice, and they push me out of the window and close the window behind me, then there’s nothing I can do about that. And that is okay, because it’s their house, they can do whatever they want inside their house. But I lose sovereignty, I lose the freedom to do what I think I want to do. And now you can argue, yeah, but that’s the nice and decentralized nature of the web, you can build your own, just like host your videos on your own server. But then I have all these hassles of hosting video for myself, or you can host your community with your own Discord server, or you can run your own TeamSpeak server or you can run whatever, on your own machines. But that’s, that is tricky to do in reality…

Gary 18:30
…and picking up on the sovereignty and making your own rules that reminds me of Tor, the Tor network,the Onion Network, which in the great scheme of things. And the idea itself is actually pretty neat. I could see that this might go in a very nice direction. But then it very fast became this. I don’t know how to say this nicely, but a cesspool of garbage. And if you want to find shady things on the internet, then you go to the Tor and perhaps the lawlessness that came to be on the on the Tor might be the result of like people making their own rules about like, what can you host on your site and what you cannot plus the privacy that Tor offers to the hosters to whoever created the websites. So that’s also a very scary aspect that comes with a decentralized web where moderation might not be the easiest, and I absolutely think that moderation should exist because well, we are humans.

Martin 19:35
Yeah, absolutely true. Nonetheless, I do think it’s tricky if the moderation is out of your hands because yes, for as long as we know most of us agree with how the moderation is done currently in most of these walled gardens, but we don’t know if that’s continuing to be the same thing and also it it hinders us in finding things. And I worry about–we hear this in the in the context of Politics a lot, that there’s like these bubbles that form. And people might fall into a trap of misinformation and then engulf themselves in a walled garden that is full of it. And it’s hard to be informed about what you’re seeing or what you might be coming across, if that is hidden from everyone’s eyes. I mean, the the privacy of a non-public community is great. Until it isn’t. And also, what if it’s a great non-public community? And I just can’t find it on the web, I have a chance of finding it. How do we find good communities if they’re not visible to the public? How do you discover these communities?

Alan Kent 20:42
To me, I think an interesting segue here is into content creators, like Google has got like the Google Creative channel now, we’ve been doing a bit of work talking about how can you get small content creators, and how can they succeed on the web. And one of the things that we’ve been hearing from them is they see creating their own website as a logical step in their growth. So they’ll start on a platform, and they’ll start on Facebook, or Twitter or whatever the platform they’re on. But they also have a fear. Like, if there’s an algorithm change, even if they don’t get knocked off the platform, if there’s an algorithm change, it can make a big difference to their traffic, and it’s out of their control. And so one of the things some of them like to do is they stay on those platforms, but they want to have their own home, they want to own their content, they want to control their destiny more. And so they’ll actually set up their own website with their own domain name, where they’ve got complete control over that content. And then they publish some of the content on the public forums. And so they can still be found, they still share some information. But then they reserved some of their content for their own community, either on their own website, or wherever it is, and so that then they they have more control, they sort of feel safer in a way because they’re less, and I won’t say completely not at the mercy, but they’re less at the mercy of these platforms and algorithm changes that might otherwise have a big impact on the amount of traffic they get.

Gary 22:03
And I think you can see this, especially on platforms like Tiktok, where they are quite tricky about what you can post, but it can bring for the Creator a ton of traffic as to the content itself. And then the creators very often move or expand to other platforms as well. Like, let’s say that a creator starts creating Vimeo videos as well, or videos on vimeo.com, or they get into Twitch, the streaming platform or YouTube or whatever. And then some, as you said, Alan, they reserved some of the content for Patreon patrons, and then that content will be only accessible to the patrons of the Creator, and then they make some money. And that’s supposedly good for them. Because you know, money brings happiness.

Alan Kent 22:53
I was just wondering, though, in terms of creating your own website, how these days, like, do you really want to create your own website?

Martin 23:01
I think these days, it’s easier than ever to create your own website, because you get to choose your own adventure, kind of like you can, let’s say you are a not so technically adept creator, and you just one photo blog, for instance, because that’s what you do photography, then I think uploading your photos onto some servers that host them for you. And then putting it into a hosted WordPress or whatever CMS you choose is relatively simple. And at least for WordPress, if you don’t care for your own domain, and are fine with like a wordpress.com. subdomain, it’s pretty much free, at least you get started for free, and then you can move from there. If you’re more tech savvy, then you can also choose to use I don’t know GitHub or GitHub Pages to host your stuff. You can set up your own server. Servers are really, really cheap these days. Cloud platforms have free quarters and free tiers that again, get you started for free. So you can start your own website on your own or under your own control without investing heavily, which I think is great. If you necessarily want to do that. I don’t know because there’s maintenance attached to it, too.

Gary 24:17
How would you think this would work on a decentralized Web? Where you–How would you create your website on a decentralized Web? Would it be any different?

Martin 24:28
I think there are multiple options for this. One is to instead of publishing it to a known destination, which is your server that is under your control, you publish it to some sort of distributed storage. I think that’s the approach that IPFS is taking, where everyone has part of the routing information and everyone has part of the content as well and thus, you have some more decentralization in the sense of if I am creating my own server, great. Now I am making the rules. Unless someone emails my hosting company and says this guy is a fraudster, or I stopped paying my bills, and then this information gets taken down. With a truly decentralized system, that would not be possible. I don’t know how exactly publishing would look like. As far as I know, publishing and IPFS looks pretty much like uploading to a web space.

Alan Kent 25:20
One of the projects I was playing with recently was using Firebase for this. I actually…using the Firebase database built in the back end, and then I actually built the whole application in JavaScript running in the browser. And so it’s using React using next JS, put it together, upload static files onto Firebase. Firebase did the authentication stuff for me with a simple JavaScript API, it did all the database access. So I just did client API calls and Firebase looked after all the authentication, and so knocked up some simple web apps quite easily, even with authentication and persistence. And there’s no code running on the server at all. Which I liked from a security aspect.

Gary 26:01
I’m sorry, I think that did you mention JavaScript? Are you saying that JavaScript is the future?

Alan Kent 26:07
It’s the present.

Martin 26:11
High five Alan, high five. That just ruined my day. Made mine, that’s so nice. No but I think that’s an interesting thing. So for us, it’s quite natural to type in URLs, at least short URLs, or scan a QR code that has a URL on it. I remember when COVID hit pretty much all the restaurants once they were open again, or every take away place that allowed me to take away food didn’t have a printed menu available, but they had like a QR code on the table so that you could scan the QR code and then see the live version because it might change on a daily basis. And they don’t want to reprint and laminate the menu every day. And I can see…so I had a coworker once, Jose Perez Auinaga. If I pronounce his name correctly, I hope I do if you listen, I’m sorry, if I butchered your name. He says he had this one post on some platform that went viral. Because he basically used a data URL that was relatively short and relatively easy to just drag into your toolbar or just like as a bookmark or something. And that was a full blown editor. So you had a tech–text editor, like a JavaScript application, if you wish, or like an HTML application, rather than anything that allowed you to edit the text file in your browser that ran from a copy paste double URL. So you could make it a QR code, no server whatsoever. It just put the HTML straight in the data URL. And that worked. And I found that very, very interesting. And I was like, we should probably explore this concept more of like, how can we encode the content into something that we can share more easily so that we don’t have to have a backend infrastructure for things that don’t necessarily require back end infrastructure? But then it’s also not searchable? Like that is only where you are at that moment? Yeah. Because there’s literally no place that can be crawled, right? It doesn’t exist. It’s ephemeral. It floats in space. That’s true. Yeah.

Gary 28:11
So in a web three world, how do you think a search engine like Steve or Google search would exist? How would that work? Because we’ve been talking about walled gardens, for example, we’ve been talking about JavaScript as a choice of programming language, for whatever reason. And it feels like search engines in general are not well equipped for these kinds of things.

Martin 28:35
That would require us to go away from the crawl, index, and serve kind of pattern that we have where we are discovering resources through URLs, and then these URLs point to documents or things that have other URLs embedded in them. And then we go there, be it a sitemap or be it just links on an HTML page, we would probably have to figure out a way to tap into where people are exchanging information. And that might be a social platform that might be a non public community, that might be a blockchain storage. Who knows, but I’m guessing we still will continue to communicate information somewhere. It’s not that these like, I can’t just create an non hosted or an hosted application that just floats in space, and then expect people to stumble upon it and then use it, there will be some form of human communication. And it could be, let’s just play a game of let’s pretend we have Steve and we want to make Steve ready for the future. We are now 20 years in the future or 100 years in the future. And there there is such a thing as local communities where people are exchanging information based on where they are, well, then this this exchange needs to happen somewhere. Maybe it’s QR codes being exchanged on phones. So in that case, Steve would have to be on the phones where the QR codes are being scanned in order to grab the QR codes as they have been scanned and then present them somewhere where people can find them. So if I’m like, Oh, dang, how did I get the information of the cafe around the corner again? Oh, I’ll ask Steve. And then Steve knows that, oh, yeah, this other person has scanned this QR code for the exam and told us it’s for this cafe, you’re searching for this cafe. So here’s the information that would be encoded in the QR code. If you were in front of that cafe, something like that. I guess, I don’t know.

Alan Kent 30:29
If you asked my point of view, I think it comes back to the content creation strategy and deciding where to publish your content. I think the creators should have control over what gets available in search engines, it sort of ties in a little bit to the value. So one of the definitions of web three is it’s going to be the value web, how do I do a better job of allowing people to get online and make a living out of it, even though they they’re sharing content and new models of getting it? We’ve talked about patron and tipping and you got concepts like product placement, and ads, and consulting and training, paid training content, there’s a whole lot of different models around for actually doing that. But I think it’s up to the content creator to decide what they want to make available to what level of community. And I think that’s the right model myself, I think they should they own their content, they should control what they are going to share and what they’re not going to share.

Martin 31:22
And I think there’s also this divide between something that exchanges value and something that is providing a value in itself as in like, if I am a service, and I want to announce my services to other people, then that monetizes itself because then people will use my services and pay for my services. But what if I am a creator, as you say, so there’s like this, this divide between people using it as a way to engage in conversations. And then there’s something where the conversation itself has value. Like, if I create a cooking tutorial, or if I create a book, or a video series, or an animated series, or a cartoon or whatever, then that has value in itself, and is not to establish value later on. So then, yeah, for this kind of for the latter kind of content, we definitely need to figure out monetization. And that means also giving creators the control and who they want to share to what level with.

Gary 32:23
And that search engines also have to figure out how to present that to users who

Martin 32:27
are looking for that. Right. Yeah, because you still want the discovery. Yeah, true. Okay.

Gary 32:33
So I think we already spent like 30 minutes on defining different versions of the web. And maybe we don’t want to dig further in without actually researching this more, because while we are not experts, we are just three people who are excited about the future of the internet. So maybe we can cut it here. And maybe in a few months, we can come back to this topic, and then discuss it further as we learn more about how the web is evolving. And that’s it for this episode. Thank you, Alan, for joining us. You can find Alan on Twitter at a Kent 99. Next time, on search off the record we’ll be carrying on with our Spotlight Series and chatting to Michelle Robbins, who is one of the most inspiring web professionals out there. We’ve been having fun with these podcast episodes. I hope you, the listener, have found them both entertaining and insightful too. Feel free to drop us a note on Twitter at Google search C or chat with us at one of the next virtual events we go to if you have any thoughts, and of course don’t forget to like and subscribe. Thank you and goodbye. See ya.

Martin 33:48
Bye Bye.