![]() ![]()
For example, we made a mega-ribbon for Word in Fig. Having an IR opens up powerful opportunities for personalizing or making application specific modifications of the UI itself. #SOME REQUIRED COMPONENTS ARE MISSING CHROME REMOTE DESKTOP WINDOWS#In Sinter, the IR is a generic XML which we found to be sufficiently expressive for encoding a majority of standard UI element types on Windows and Mac. The Sinter proxy relays keystrokes and other user inputs back to the scraper, and the scraper relays incremental changes in the UI back to the proxy. In Sinter, a scraper on Windows mines the Windows-specific UI model of the application, and converts it into Sinter’s IR which is shipped to a client system–Mac in this example, where a proxy converts the IR back to a native representation of these elements that VoiceOver understands. #SOME REQUIRED COMPONENTS ARE MISSING CHROME REMOTE DESKTOP MAC#The figure corresponds to a scenario where a user wants to run VoiceOver, the Mac screen reader, to read a Windows application on a remote system. 1 is a high level schematic of this idea. A screen reader on the client system can then read the native rendering of the IR. The key idea then is to create a UI model of the application’s GUI from the existing, platform-specific accessibility APIs, analogous to an HTML document object model (DOM) tree, convert the model into a generic intermediate representation (IR), and finally render the IR encoding of the application GUI on a different platform, using native UI widgets. Sinter is predicated on the observation that applications on every OS consist of similar User Interface (UI) widgets such as buttons, drop-down menus, text fields, etc. Our Sinter system provides such a solution. Such scenarios are the norm these days and, thus, calls for platform-independent remote access solutions for screen reader users. Similarly, a Windows desktop user may use VMware workstation to develop and test a Linux server application. For instance, a Mac user may access a cloud-based Windows remote desktop to run an application required for her job. Platform-specific remote access solutions are clearly inadequate in the modern computing era where computer users increasingly use applications designed for different OSes, spanning desktops, laptops, tablets, mobile devices, cloud and other platforms. Consequently, NVDARemotelike solutions require both the remote and local platforms to be similar. The serious problem with such a solution is that screen readers are locked into a single operating system platform the differences in the underlying accessibility APIs (e.g., Microsoft’s MSAA and UI Automation, Apple’s Accessibility, and GNOME’s ATK & AT-SPI) has been a strong barrier to portability. Getting this to work requires identical screen readers, one running remotely and the other locally. A second approach, exemplified by commercial systems such as NVDARemote and JAWS-Tandem, is to intercept text from the remote screen reader just before audio synthesis, relay this text to the local client and synthesize audio locally. But network delays, especially over WANs, can introduce unacceptable latencies to relaying audio, a fact validated by our experiments. One obvious solution is to run a screen reader on the remote host and relay the synthesized audio to the client. Screen readers require semantic information, such as text and hierarchical relationships to narrate screen contents-all of this information is lost by the time the screen is rendered as a bitmap. These users rely on screen readers to interact with digital content. Unfortunately, remote desktop technology in its current incarnation, namely via emulation of the graphics frame buffer, simply will not work for blind users. Remote desktop access is indispensable to users engaged in activities such as telecommuting, distance learning, remote troubleshooting and maintenance. ![]() This approach yields seamless access to remote applications, as if they were running on the local desktop. Virtualization emulates the graphics card frame buffer in which the pixel values are scraped from the screen of the remote system, and redrawn as a simple bitmap on the client’s local screen. ![]() The graphical display of the remote system is virtualized and shipped across the network to the client. ![]() Remote desktop technology is primarily used to access applications that are hosted on remote machines, such as a physician accessing a medical records application running on an office server from her home. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |