[LispM] Introduction

[Earl DuCaine]

 >> My apologies to the authors of ld: my hacking has been a dumping
 >> ground for various ideas with no thought of being able to reintegrate
 >> them back into the main branch.  Very rude of me.
 >
 > No apology needed, as long as you don’t get upset when the next few
 > changes really screw up your branch. (I’m redoing all of the logging
 > so that it can be configured by source and level, the changes are
 > going to be quite invasive)

Maybe that will encourage me to brake my bad habit.

 >> - SDL2 ignores your local xkb (X11 keyboard mapping.  Very
 >>   inconvenient if you've using a Dvorak keyboard.  SDL1 works just
 >>   fine however.
 >
 > I’ve never met or used a Dvorak keyboard. Is it actually a different
 > keyboard with different firmware or did you just rearrange the caps
 > and mapping=3D3F Does SDL think “A” is some other key or does it just
 > not know what you pressed=3D3F

Actually, I use a totally conventional keyboard, just a variant layout
configured at the OS desktop level.  (Oddly, I touch type using Dvorak,
but I have no visual association of character to keys, so on mobile
devices with a visual keyboard I have to use QWERTY. In fact, it never
occurs to me that the two layouts are different until I start using a
new device and inadvertently configure the keyboard to Dvorak, which is
impossible for me to use efficiently)

In modern X11-type systems the Dvorak mapping file can be found at:

/usr/share/X11/xkb/symbols/us

It's distributed in human readable form and dynamically (re)compiled as
needed. I use a few custom modifications to make Lisp easier to type --
thus I had to 'undo' the ld's remapping of ( and [

My knowledge of SDL is still defective, so I can't be sure the issue is
really inherent to SDL2 (I suspect it's not) or is just incidental to
the way SDL2 is setup in ld.  Roughly, the immediate cause is that
within ld, the SDL2 keyboard mapping utilizes scancodes, eg

map[SDL_SCANCODE_W] = 0062; // W

while the SDL12 mapping uses ascii character codes

map['W'] = 0062; // W


 >> - TUN/TAP was difficult for me to set up. (my own ignorance) Still
 >>   haven't figured out a way to run it conveniently without resorting
 >>   to running the emulator under root permissions.
 >
 > When creating the interface: sudo ip tuntap add dev (name) mode tap
 > user (user-name)
 >
 > The given user then has access to the device without having to be
 > root.

I'll give that a try. My virtual network environment is bit complex
because I'm using a separate network namespace. (I kept disabling my
local Ethernet, tcp/ip network stack whenever bridging was activated to
the virtual device.) So, I'm guessing I'll have to make a few other
modifications as well.