Let’s start talking about web products. I’m really talking about all kinds of products and the taxonomy of those products. That means we also have to talk about web technology, which is basically the evolution of the web itself.
Now, I’ll say my advantage here is that I’m old. I lived through Web 1.0 all the way to now. On one side, we can talk about this as a history, which is great—except that then we tend to believe that anything that doesn’t embrace the latest technologies is somehow bad. Well, except that sometimes it just doesn’t need those technologies.
So the question becomes: is this a way of classifying web technology for different kinds of web products, or is it an actual history? I kind of want to say it’s a little bit of both. We can look at it in the context of history, but it’s much more useful when we look at it as a taxonomy—as a way of classifying what a product actually needs.
Now, to make this even more confusing, we also have this idea of Web3. You’re like, “Well, there’s Web 3.0 and there’s also Web3. Why?” Because they really wanted to mess with us. A lot of this stuff is really just marketing terms. Web3 has basically nothing to do with the technology of the web itself. It’s about blockchain, crypto decentralization, NFTs—stuff like that. We’re not going to cover that specifically because it doesn’t really fit into what we’re talking about here. It’s kind of a technology off to the side.
At the very beginning, the technology is pretty basic: using the web for file transfer.
Think about something like a class web repository. There are lots of examples—just examples—where the point is simply file transfer. That right there is a web technology. I sometimes call it Web 0.5. File transfer is one of the underlying capabilities that the web supports.
Now, you might say, “Well, I’m providing these examples. I should turn this into something fancier, something more technologically advanced.” But it doesn’t matter. I just want to send you the file. Why add all this other stuff when the goal is file transfer?
That website, that web product, is Web 0.5. That’s the technology it’s using, and it doesn’t need anything else. This is where I start to separate products from technologies. There are products defined by what they need, and there are technologies defined by what capabilities they offer. Once you look at it that way, you can classify this kind of activity as Web 0.5.
After that, the content web was much more about presentation than anything else. Then we moved into the social web, where there was a lot of content sharing. We’re not going to get into the social web.
Then there’s the mobile web. What’s interesting here is that when we see all the JavaScript stuff, we think, “Oh, that’s client-side programming.” But why do we have client-side programming in the first place? It’s because of mobile devices.
What do mobile devices deal with? DIL networks—disconnected, intermittent, and latent. That means I want a lot of the programming on the device itself rather than on the server, because before this era, all the programming happened on the server. Now we say, “Some things should live on the client.” But we’re not doing that just because it’s fun. We’re doing it for a specific reason. The technology changes because the requirements change.
Next comes personalization. There are personalized search engines now. You all can search for “python,” and I can search for “python,” and we might get different results. Some of you might get results about the snake. I’ll probably get the programming language. My Aunt—who I always pick on and who’s not very technology-focused—she searches for “python,” she’s definitely getting the snake.
Why am I getting different results than my aunt? Why do we all get different results? That’s personalization, and that’s one of the defining aspects of Web 3.0.
So if you tell me, “Just put keywords on your site so people can find it,” I’ll say, sure, that was great—for Web 1.0. That was when we cared a lot about keywords, commerce, and search engines in a very literal way. Now it’s about topics and topic modeling. It’s about the relationship between what the user wants and what the search engine decides to provide.
Beyond that, we start talking about personalized websites—and that’s what we’re going to build. We’re going to take a Web 3.5 approach, focusing on personalization at the website level. That means we will need programming. Not a ton of programming– definitely more thinking than programming.
As it turns out, search engines tend to lead the way when it comes to new technology. For us, that means we’re going to focus on the machine learning aspects of personalization. If you keep going forward, you get deeper into AI-driven personalization. We’re not going to build AI directly into our product—but we will use AI to figure out how to build it.