Here are my thoughts.

Q: Will social media companies let go of branded buttons?

Honestly, I don’t expect them to. But that is not a problem, I expect they will become a sink for the intent by providing an tag in their page and parsing it. And that is enough to get the process started.

Q: Users will never get the concept

If implemented correctly users will never see “intents” or “activities”, they will just see a “share”, or an “edit” or “pick” button and a list of their services that they can use.

Q: Why don’t we just let Facebook/Twitter dominate – do users really want choice?

It is more than user choice, it is about the developer not having to explicitly support new networks, or removing old dead networks from their code.

It is more than just “share”, which undoubtedly is the biggest first use-case.

Q: In page UI vs chrome UI

You are correct, it is up to the UA to decide, however I am pretty confident and I know in Chrome’s case (you can check the commits) that it will be outside of the Page UI and away from the context of the click. We need to make sure that all of this is un-spoofable; the second that it is only in-page then it becomes open for attack via spoofed code.

Q: Common iconography and verbs

I believe this is important, but can be managed outside the spec. I would like webinents.org to contain the de facto set of common verbs and have activity-streams to maintain their set (and objects), but also let the wide community define their own verbs.

We have a basic mapping to “widgets” (http://widgets.webintents.org) that help define the common look and feel, but I am very very open to this being contributed to via 3rd parties as it is outside my skill-set… In fact github.com/PaulKinlan/WebIntents is waiting for pull requests

Really like the idea of a wizard and is part of what I intended widgets.webintents.org to be.

Re your AddThis mention… Web operators do look to folks like us increasingly for better insight into the effectiveness of these tools, via analytics and other services, and help optimizing these kinds of experiences by offering appropriate and effective choices for users. While not all sites use them in this way, our own tools for example can automatically personalize options for individual users, present the tools that make sense for them, and give the operator metrics on all aspects of that (and a lot more to boot). And they can do all of this via APIs, on top of which they can build whatever UI they might want, including those that only include 3rd party buttons. Even in the simplest case of the ubiquitous “drop-down menus”, the options that are presented to users are based on extensive data analysis (see a glimpse). So there’s more going on here than may meet the eye initially.

On top of that, though, I’m psyched about efforts like WebIntents and frankly have been wanting these things to go forward even more quickly for some time now (we’ve been advocates for as long as anyone). The open stack should clearly encompass the types of core sharing operations we’ve been observing for some time — and our advocacy on things like OExchange, XRD/-based service-discovery on hosts for sharing, and the like, I think shows the support. While the UX angle is the most obvious, there are other implications for operators that should be part of the conversation as well, and that’s where we’re excited to help all of this push forward.

In any event, keep up the great advocacy, see you on the lists!