From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from dc.cis.okstate.edu (dc.cis.okstate.edu [139.78.103.93]) by speech.braille.uwo.ca (Postfix) with ESMTP id C884C102EA for ; Fri, 15 Jan 2010 09:46:14 -0500 (EST) Received: from dc.cis.okstate.edu (localhost.cis.okstate.edu [127.0.0.1]) by dc.cis.okstate.edu (8.14.2/8.13.8) with ESMTP id o0FEkEdd036451 for ; Fri, 15 Jan 2010 08:46:14 -0600 (CST) (envelope-from martin@dc.cis.okstate.edu) Message-Id: <201001151446.o0FEkEdd036451@dc.cis.okstate.edu> To: "Speakup is a screen review system for Linux." Subject: Full screen Presentation (Waswhere am I on the screen? ) MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-ID: <36449.1263566774.1@dc.cis.okstate.edu> Date: Fri, 15 Jan 2010 08:46:14 -0600 From: Martin McCormick X-BeenThere: speakup@braille.uwo.ca X-Mailman-Version: 2.1.12 Precedence: list Reply-To: "Speakup is a screen review system for Linux." List-Id: "Speakup is a screen review system for Linux." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 15 Jan 2010 14:46:15 -0000 Before I started using speakup right around Newyear's Day, I was using a screen reader I wrote in assembler, starting in the very late eighties. I seem to remember that some forms of ansi.sys in DOS would produce multiple prints of whatever was being sent to the screen. Do any of you remember hooking interrupt 0x10 better known as the video BIOS interrupt? I think the multiple prints consisted of one valid character of the desired color plus another just before or just after that had the same foreground and background attributes making it invisible on the screen for sighted users, but you could sure hear it if you were ignoring attributes and just catching the ASCII value in the lower 8 bits. What we are dealing with now is more likely the same problem. I used 3ansi.sys to get sensable output in general DOS and MSKermit as the VT100 emulator which normally gave clear output. I always think it is best to get readable output as it comes in whenever possible and MSKermit did cause the FreeBSD installer and similar full-screen programs to behave fairly well as long as the output from the program came in sequentially. What I hear now is a slow and dragged out presentation as a new screen comes in and a nice snappy normal presentation upon pressing the keypad Plus to review the screen. Speakup is simply trying to read what it is receiving which I believe is a lot of double-printing as the terminal emulation paints a new screen. I guess it comes down to figuring out whether existing tools can be configured to produce clear speech as new screens come in or if I need to get busy on some sort of filter. I think it was more dumb luck than anything else, but MSKermit's VT100 emulator did a great job of handling the installer program's output. I've got about 8 systems to do this installation on so everything that makes it smoother is greatly appreciated. So far, the change from my old system to speakup has been fantastic. The problem with the full-screen installer and VT100 emulation is the only thing that, so far, used to work better and is pretty rough at this time. I don't think the FreeBSD sysinstall program is doing anything that out of the ordinary but it probably uses a lot of curses-based library routines for control. Trying to decipher the sound and fury of the screen output has prompted a few curses of my own. Hmm. Is this still the same screen again or did I go to the next screen? I think it said something about rebooting. Is this what Martian sounds like? Martin McCormick WB5AGZ Stillwater, OK Systems Engineer OSU Information Technology Department Telecommunications Services Group