os
ZXNet echo conference «code.zx»
From Arseniy Astapenko → To All 20 April 1998
* Forwarded from BBRZXNet
from : SAV/D.M.G./U.A.
to: Raskazov Sergey Aleksandrovich
subj : Approximate view of our microOS.
date : 11-04-98
time: 17:43
Hello Sergey!
I thought about this idea a little more and added a couple of
thoughts. So:
1. What is it?
This is a shell, not an OS based on Tr-DOS
(like Windows itself). Disk with files of this
the shell is no different from the usual one.
The core contains 128 kB of RAM, i.e. unlike Is-DOS,
this shell represents more resources (brains, access
to I/O devises), in addition to this, a powerful device for accessing
screen file and disk. It is quite possible that there will be
Disk polling is not part of either Tr-DOS or MS-DOS.
Access to the screen will be very sophisticated, i.e. everything
procedures for displaying characters, printing sprites, windows, lines,
points and in general all graphic objects that are
basic (standard) will be filled into Bios.
In the future we can see the creation of emulation
multitasking, i.e. running multiple processes
at the same time. This can be interpreted in two ways:
a) everything works simultaneously, on three levels
access:
I. Prohibiting everything for the duration of its execution
other processes (formatting, disk operations)
II. Working in multitasking mode, i.e. onethe interrupt works one way, the other works differently, etc. and then
everything repeats itself. (Most processes)
III. One-time, called every interrupt (clock,
music, key polling, etc.)
b) All processes hang in memory, but only
the process that is current, i.e. moving into
another process you stop the previous one (but not
destroy it) and start a new one. Moreover, when
returning to the old process, you continue working in
it from the moment where we graduated, and how
as if you never left it.
2. What does this give?
This shell will allow you to use quite effectively
your resources (both computer and brain), because
the ability to 'simultaneously' work in several
processes will provide high speed and greater
return. Next is the ease of downloading. Due to the fact that the core
is TR-DOS, the shell loading speed will not be
exceed the boot loading speed, so switching from
shell to TrDos and back will take a minimum
time.
Simplicity and freedom of creativity.
Obviously, the Bios will be configured to execute files
: com, exe, bat (meaning analogies). Those.
programs can be written in the form of a bat file,
using bios resources and com swapping and
exe files.
In theory, this shell is based on the principle of convenienceobject-oriented programming, i.e. all this
crap is aimed at solving a specific
object-task, and so that this solution (writing
program) was available to everyone.
3. For whom is this needed?
Actually, this is intended to resuscitate life in
Bobruisk. If every user is able
write an executable file, then our connections will be
filled not only with soap, but also with programs.
4. Alternative...
Of course there is an alternative to this project - perhaps
creating a design program using a tricky method: this
a program is a piece of language, only more
aimed towards specialization in
user tasks.
In my opinion, what I said above is not clear.
So, let me explain: At the input there is a text file with the so-called.
'program' [actually call it a program
under no circumstances]. The program compiles this
file into an executable file.
Now about the 'program':
There are actually no operators in it - operators
are the names of tasks launched together with the file,
eg : Viewtxt Help.C, Call MAIN+.C, Load ANTI.C, Run
boot.B. But these operators are not really operators, they are names
tasks. But operators are also provided: Print, Cls, For
etc. There are also functions: Inkey$, Len, Sqr, Sgn and
etc. In theory, this is a highly specialized microlanguage, but inwhich only has a compiler.
5. Choice!
In truth, all projects (Language, shell,
designer) are very important and necessary. I want
do everything, but, unfortunately, I have neither time nor
physical capabilities to master so much
grandiose projects.
This raises the acute question of choosing the project
which I'll be poking around this summer.
Once I filled out my program of action, but since then
so much has changed that we have to do it all over again
transfer everything from the beginning.
6. Action program:
1. Add every little detail (CGA grafix editor &
other)
2. Write a MODEM romp poker (ex-strip).
3. One of the super grandiose projects:
Language - Shell - Designer - Interactive game -
Step-by-step super strategy with self-developing
intelligence (with the ability to support a modem option)
A huge number of small programs, utilities, converters and
etc.
This is my grand plan for the near future!
That's all for now, it's time to burn it to disk, otherwise...
With the best wishes, SAV.