 The best of being here is to side the low ground compared to everyone else who's dead. I'm a blogger, until very recently I was a dirty tech blogger. And now that that's over with, I'm trying to figure out a better way to get, to have web journalists and web journalism serve its readers and listeners and viewers rather than harm people. And I think that a web-wide annotation platform like the One Hypothesis is building is a big part of that. This is basically why I think that the web itself needs a place to respond to content. These little buzz-worthy things on the side are sort of my philosophical reasons for that. I think that the incentives get messed up by the motivations of publishers themselves. And the other ways that they've allowed readers to respond. And I've got sort of four use cases for this that I'll go through that explain why. And I've sort of, I've made up a new one that I'll share at the end that is sort of my dream of what a tool like this can do. The first problem that I want to talk about is comments, which everybody knows are terrible things. The problem is trolls. That's the colloquial term for people who just like to stir things up in the comments rather than constructively respond to things that are published on the web. And one problem that I noticed every day working as a professional full-time tech blogger was that commenters are not incentivized to read the material that they respond to. So my first thought when I first understood what a tool like Hypothesis could do was that if you had to point to a specific passage, to a specific object on the page, they have to at least know what that sentence says, right? So that's a pretty simple use case. If you're able to view at a high level all the annotations on the page with a heat map or something, that drives attention to the important parts of the content, which is easier to deal with than the whole thing at once. And on top of all of that, what you end up with if you've got very specific, good-threaded commenting on an entire page worth of stuff is a goldmine of sentiment of responses. As a publisher, you've got this goldmine of data about what worked for people and what didn't. The next use case I want to talk about is fact-checking, something that was entirely absent from the kind of news that I was doing before. The reason is because the cost of doing it is really high. It's an entirely manual task. Often it requires an entire site dedicated just to doing that, like Politifact, which is where I got this awesome graphic. This is what they put on their response to you if your article is entirely full of lies. I feel like annotations can provide a standardized way of tracking reputation, one that can compare across publications and can compare across one publication over time, can follow one author around to his or her various places. One thing I observed all the time as a full-time blogger was that people move around a lot and it's really hard to track their reputation across domains. That's one of the reasons why they move, in fact. So this is why a separate web layer for fact-checking might help. Discovery is another one that I think... There are two ways that web publications are driving Discovery right now. One is with third-party little widgets that go at the bottom of the page that are uniformly awful because they're designed to optimize for bad click-driven metrics that add support to publications. Publishers themselves are not incentivized to drive readers away from their site to other places when their business models are ad-based. So I figure a web annotation solution can allow authors themselves to annotate the links in their posts, making the research that they've done more discoverable if they're so inclined, but readers can do that as well. One thing I've thought a lot about is private versus public when it comes to these kinds of annotations. I think either one could be good. If you've got a small closed network who's working on something and you can send people to other related content that way, that's great, but if you've got a huge social network that anyone can subscribe to and find articles that you thought were interesting from something that you're all looking at, I think that's great. The hardest problem right now that I think web annotation can solve for journalists is attribution, which is basically entirely voluntary right now on the web. And so with a separate web layer for annotations, you can graft attribution back on to a page if it was left off. That's really important. The author can do it himself or herself, or a reader can do it, or some other third party can do it. So those are basically the four ways in which a separate web-wide layer of annotations can help with the existing work that web journalists do. For fun, I invented a new way of using a service like this that I hope a lot of you are thinking about, and I even made up a nice buzzword for it that I don't think anyone's using, called hyper-tagging, which I think gets at a lot of these different problems at once. And if I were good, I would have done a mock-up of what it looks like for you, but hopefully these bullet points will explain. My idea of a hyper-tag is a bundle of links that are joined by web annotations. So the key to it is that all of those links can be discovered on any one of those pages by viewing the annotation that you're looking at. And ideally, there would be this descriptive text on the hyper-tag in the annotation itself that would be addressable at its own URL, where the person who creates the hyper-tag can describe the relationship. And then that becomes a sort of curatorial, meta-journalistic project where people who are interested can create meaning between different articles and web objects and connect to dots. So those are the ways I've come up with so far for journalists to use a tool like this. And I'd love to talk about more sometime. Thanks.