Hi Mateusz, I agree it is convoluted, but it's fun to play with such things. The user experience is less than optimum because NVDA has its own ideas about what part of a terminal window has changed between consecutive states. So the virtualization approach is not off the table and is, indeed, scheduled for tonight. Working in such environments is just a matter of training. No outstanding memory or very high IQ or any such thing is required. Rather, when working this way on a daily basis, your memory would gradually adjust to this kind of demand, just as it naturally learned over time to recall points of interest when driving, if indeed you do drive. Our brains wire in such a way that we develop an encoding which is isomorphous to the aspects of reality to which we relate, which is the exact same process as learning a language. You might go so far as to say that reality itself is a self-processing self-configuring language, but now that's metaphysics and way off topic. Best, Felix
Am Mo., 16. März 2020 um 14:46 Uhr schrieb Mateusz Viste <mate...@viste.fr>: > > Wow, that is much more convoluted than I was expecting. I'm impressed > you were able to figure out such a solution. > > Is this NVDA-over-ssh-over-dosemu approach providing satisfying > usability compared to a native DOS screen reader like JAWS or ASAP? In > other words, is there any point in trying to run a native screen reader > within FreeDOS? As a test I tried this morning using FreeDOS being > blindfolded, using only JAWS + espeak, and I am staggered how you guys > are able to do anything at all in such environment. I was barely able to > open a text file, and failed miserably trying to edit it - all this seem > to require an outstanding memory just to keep track of what is supposed > to be displayed on the screen at any given moment. > > Mateusz > > > > On 16/03/2020 14:20, Felix G. wrote: > > Hi Mateusz, > > I logged into a Linux server via SSH from Windows, installed Dosemu, > > then started Dosemu with the -t option, putting it into terminal mode. > > In effect, my Windows screen reader, which is called NVDA, detects and > > reads changes to that terminal session, thereby giving access to the > > DOS environment. > > Here is how I got Dosemu2: > > $ sudo add-apt-repository ppa:dosemu2/ppa > > $ sudo apt-get install dosemu2 > > I then downloaded the FreeDOS Kernel, extracted KERNEL386.SYS, and > > copied it to ~/.dosemu/drive_c/KERNEL.SYS so Dosemu boots it up. I > > also replaced Dosemu's command interpreter by the COMMAND.COM that > > comes with FreeDOS, and then I tweaked autoexec.bat and fdconfig.sys > > until there was not a single error message during boot. Does a rather > > convincing DOS approximation. Since Dosemu's virtual C: drive is just > > a plain old directory on my Linux server, I can just unzip games to it > > and start them up, without so much as exiting Dosemu. > > Best, > > Felix > > > > Am Mo., 16. März 2020 um 14:03 Uhr schrieb Mateusz Viste <mate...@viste.fr>: > >> > >> On 16/03/2020 13:12, Felix G. wrote: > >>> In the meantime I was able to play the Time And Magik trilogy by Level > >>> 9 in Dosemu2, using the current FreeDOS kernel 1.2 > >> > >> May I ask how you achieved this? Have you managed to install a screen > >> reader within DOSemu, which would talk through some Linux TTS? I'm > >> geniunely curious. > >> > >> Mateusz > > > _______________________________________________ > Freedos-user mailing list > Freedos-user@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/freedos-user _______________________________________________ Freedos-user mailing list Freedos-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-user
