I have just built a few pages to understand the user experience of Web
Intents/Activities using Pauls shim.
I picked the most well understood use case, sharing a URL and created
a number of proxy pages to bridge-the-gap to real services. The proxy
pages collect the Web Intent JSON objects and maps the data into the
correct query string to call the required service.
Here are some of my first thoughts from trying this:
All the services accept metadata about the page, typically this is 4
properties URL, title, description and tags. Without this additional
data you end up providing a reduced experience, as they pre-populate
the form provided by the service. At the moment it’s easy to change
Paul’s code to exchange an object with key/value pairs. I think this
approach should be added to the specification. It may be worth looking
at oExchange which tried to address this very issue.
Precise terminology for a verb
The more precise a verb the stronger a call to action it creates.
Bookmark is not currently a verb specified on webintents.org
. I could
have used the verb ‘save’ rather than ‘bookmark’ for saving a URL, but
the action is called bookmarking as far as I am concerned. The more
generalist term, ‘save’, is not strong enough at UI level. I think it
would be good to user test the verbs or use focus groups to determine
the right terminology. Tantek Çelik and Erin Jo Richey have started
running sessions around this area. I am going to try doing one at
Mime type text/uri-list and Huffduffer
The Huffduffer example brings up another issue, technically we are
sharing a URL, but for the service to work correctly the URL has to
point to a resource that is a mpeg file. It maybe that we leave the
service to discover the type of media provided by the URL.
More importantly I am not sure I would expect the Huffduffer example
to be listed with the same service that share URLs in general. At the
moment I am not sure the current specification allows for an
acceptable UX for the Huffduffer example.
I am going to keep on building around the UI ideas. The next step is
to exchange the intent picker window for an in context pop up menu.
Then move on to creating an in-line service interface.