Many sites offer search functionality to users. Often these sites include some type of external search provided by the sites preferred search service.
Search infrastructure is notoriously difficult to integrate within sites, either as a plugin or your own custom service.
Instead, this could be delegated to the users preferred choice of search engine.
There are lots of services that offer the ability to print letters, documents, business cards directly from inside an application and have them physically delivered to you.
Most of them need a print driver to be installed. On cloud platforms and other devices there is unlikely to be any way to install a driver. Web Intents could be used to get offer this facility and deliver the file to the client.
I want to be able to build systems where I can print something anywhere and all I have to worry about is making the print button look nice.
When you upload a photo to picasa, there is the option to edit the image. It loads up Picnic. I am not particularly keen on Picnic.
Every image service should be able to offer the ability to edit images without having to perform complex service integrations.
I imagine the ability to quickly remove redeye from any image that I push to an application. For example, twitpic could quickly offer this ability with little or no service integration.
This isn’t a problem with what Stack Overflow have done, but rather the problem inherit with an Open Federated authentication system. Namely, how do you support all the services a user might use.
In this instance I am lucky, I use one of the options presented as my authentication provider, but if I don’t then it is pretty hard to work out what to do (which is why I think they have a “click here to sign up” link)
Web Intents could provide a way to only show to the user a list of services that they use that support authentication. Client application never has to know the entire list.
If you own a feed that is hosted via feedburner, it is possible to subscribe to email alerts via the service.
It should be possible for users to choose their subscription service of choice so that they can get notified of updates to a feed in a manner that is fit for them. This will also help developers because they don’t have to integrate with a entire plethora of different subscription providers.
A well defined list of potential intents that services can use as a central repository for discovery and documentation.
The interesting thing is that the XML Schema defines a unique URL for each specific verb. For example http://activitystrea.ms/head/activity-schema.html#image defines an image that includes specific meta-data over and above that which might be defined in file. This meta-data is also a lot more managable for service applications as it can be defined in JSON.
Would love to see “subscribe”, “purchase” and “reserve”
Need to speak to Messina et al.
Share and bookmark are often conflated - in this case there are at least 120 different services that use a combination of share and bookmark. The Web Intents system will be able to distinguish between both because the decision is made by the service provider, in some cases I expect a service to offer “bookmark” and “share”.
This is one of the primary uses cases for Web Intents and should obviate the need for this API
