4 ms·
From the feature list: "Can request access to one or more advanced and/or privacy-sensitive capabilities" and "Can receive notifications from the cloud." GMai
by ShabbyDoo 16y ago
From the feature list: "Can request access to one or more advanced and/or privacy-sensitive capabilities" and "Can receive notifications from the cloud."
GMail, etc. has already proven that the browser provides a rich enough UI experience to replace desktop apps, but the general lack of integration with the services of an operating system still forces me to either (1) use desktop apps for some things or (2) go through convoluted processes.
An example: I used to work on a large photo sharing site, and making it easy for non-technical users to upload photos was important but difficult. A desktop app has the advantage of being able to recognize connected hardware, interrogate obscure file structures (DSIM blah, etc.), and show the user a pretty grid of photos from the camera -- all triggered on the hardware connection event. Recently, I told my dad to use Picasa as an uploading front-end. His eyes glaze over when I have to tell him how to navigate a directory structure, so a site which requires HTTP Upload is DOA for him. HTML5 will provide multi-file select, but the degree of available automation is terrible when compared to the desktop.
Here's what I'd like to see: A set of standard, OS-agnostic profiles for services provided by an operating system and common peripherals which would allow web-based apps to do things like slurp photos from a camera, print a pixel-perfect document, or capture a scanned image. Operating system vendors (or 3rd parties) could wrap OS-specific driver profiles with these Javascript/Browser-accessible APIs, and users could grant certain domains the rights to use them. Imagine a TWAIN driver wrapper for Windows which would allow a webapp to initiate a scanning request and capture the image without the user saving the file to disk and then uploading it.
The proposed notifications service is kind of like this except that it is designed to receive notifications from a bunch of sites even if no webapp is currently running.
I think the search service design should be inverted. There should be a standard search service interface which sites could implement, but the local operating system should implement this same interface and should be queried in parallel. A user could search for photos on his photo sharing site, a Dropbox-esque site, AND his local filesystem. But, the search service itself could be SaaS. Why make the desktop heavier?
What about a dialer/voice profile for mobile devices (and desktop as well)? What about a "screen management" profile which would allow a web-based Powerpoint competitor to automatically show the presentation on the projector but the slide notes on the laptop monitor? What other profiles would be interesting which could reduce the need for applications to target specific operating systems? What about a "headless" profile and a "system tray" profile so that Pandora could be implemented using a combination of HTML5 audio, an icon on the system tray, and the notification service for song titles?
- ShabbyDoo 16y agoAs I re-read my submission, I thought about ActiveX controls circa 1999. They certainly could have been used to solve such problems (at least on Windows), but the ability for any old site to install native code made them a security disaster. What's different here? The entity you trust to provide you with a secure browser is the same entity providing a set of profiles, and you would opt into which sites could access these profiles. And, there's no "run arbitrary code" profile.
- kijinbear 16y agoThere's still a problem. Most users would just click "Yes" whenever a web site asked for access. Chrome addons routinely request access to "all files on your computer" and very few users complain about it. Fast-forward, and it's ActiveX all over again. Do people know better these days? I'm not sure.
- ShabbyDoo 16y agoI wonder if the security concerns could be limited by opt-outable warnings on first use. Let's say you gave a malicious site access to the camera profile. Would it be sufficient for an implementer of the profile to show a dialog saying, "Click here to upload all your photos to XYZ.com"? [Or, even better, show thumbnails of the photos which would be uploaded] Is it more real for users to allow specific access at a specific time than just a blanket "ok" at install time? Is there a way to have one's cake and eat it too?
- DjDarkman 16y agoIt's very hard to get these right and not having an annoying nag-ware application.