Mail Archives: opendos/2001/01/04/15:34:25
Hi all,
First, I´d like to wish everyone listening to this forum a happy New Year.
Given the long silence after my (rather pessimistic) "essay" in the past
millenium, I am assuming that you either were shocked by my provocating
statements or just by the fact how much the average consumers´ privacy is
confined already - often even without their knowlegde and explicit consent
(when speaking particularly of Internet users using Windows, default
settings are just soooo convenient to leave them as they are: Ever wondered
what this nice sounding "Control panel"->"Internet options"->"Extended
properties"->"Enable profile assistant" actually does? Knowing the
side-effects of the Internet search function in the toolbar, and
web-based auto-link-completion?... ;-> Well, I know, most of you are
experienced to avoid such issues, so this probably does not apply to
you (we use DOS, don´t we?!? ;-), but I guess the sad story is the majority
of PC users today has literally windows and gates (so to speak ;-) open to
abuse, providing info to interested parties in the hidden "public" they
would not even tell their neighbours...
So, maybe there´s a slight hope, that the silence was because you all
feel similary threated by the mentioned combination of closed-sourced
technologies & sometimes strange ideologies, in a world which is moving
faster every day, where short term issues count and only money seems to
talk any more, and - even more important IMHO - that the consequences
of using some of the key technologies are unfortunately no longer fully
understood by most, probably due to lacking higher technical (and social???)
education, paired with marketing controlled Zeitgeist ("fun generation") and
simple minded or even blind belief in technology, and last but not least by
intended obfuscation on the other side (with sometimes only marginally
developed humanitarian conscience). Who takes the responsibility?
But fortunately there are alternatives (there always are, if one takes
the time to look for them). (Ironically, having said this, I´m "accidently"
sitting on a Windows box with Outlook Express, once more my impression
is: pretty at the first glance, but with no real substance. And, yes, I´m
already close to ruining my eyes and getting sado-maso phantasies due
to missing essential features and its many unergonomic design
details... ;-> Know your enemy, it seems... ;->)
We have Linux and GNU software, which is driven by alternatives and is
evolving faster than ever thought, and we still have DOS, which is still
powerful and - with the right tools in reach (like, to mention some of my
personal favorits, the 4DOS command shell, the DOS Navigator file manager,
the TSE & QEDIT editors, the EmTeX word processing/publishing system,
the Arachne web browser, FreeKEYB keyboard driver & console extension) -
convenient enough for many task, but, although far from being "dead",
unfortunately is (and will be) more and more stagnating - until either the
FreeDOS kernel will become much better than it already is, or the more or
less complete sources of a commercial DOS like the technical superior
DR-DOS would be opened up to let the community continue what was
started 25 years ago and brought forward over the years by a group of
brilliant developers (and some visionars) at Digital Research, Novell,
Caldera, and other parties.
However, the backbone of a community as we need it are many people who
are willing to support such a project in every possible manner (not only in
this forum) by providing expertise, infra-structure, time, good will and an
open mind mixed with some sort of evangelism and the necessary persistence,
and, yes, sometimes money (to grease the "interface" to the outer world ;-).
Personally, I am seeing a possible DOS (and thereby DR-DOS) future in
the following areas. Well, rather unsorted, but I´d still like to hear
*your* estimation:
- (for up to 10 years and longer) Low footprint and fast OS in many kind of
x86-based embedded systems (especially if the kernel would be made
even more configurable and modular - it currently requires 128 Kb
to boot - and if the multitasker would get better real-time functionality.
Also helpful would be improved serial communication support under
TASKMGR.)
- (for max. 5 years) Advanced file by file replacements for similar files
and tools in other DOSes, Windows 9x, ME, NT, 2K, or OS/2 (way too
many ideas to list here, but the approach is "best of all worlds")
- Various kernel improvements (too much too list, but besides others these
things are on my wish list to better support current systems: 64-bit LBA,
FAT32, LFN support [a) native VFAT compatible plug-in *and* b) a loadable
"transparent" virtual FS model to work seamless with other DOSes, e.g.
using 4DOS DESCRIPT.ION files to store the info instead of polluting the
directory entries], full DOS 7.10 API plus extensions, improved memory
management. (Now, that Microsoft is finally dropping DOS, this also has
something for good: 100% compatibility, which often prevented brighter
solutions, is no longer the top issue. Once having laid a rock-solid base
developing a DOS 7.10 compatible OS, we would be free to design and
implement the final "DOS 9" according to our own wishes.)
If the kernel should still be useable on *new* *desktop* hardware as a
*native* OS in a few years, USB, Firewire etc. support will be required,
as the traditional PC architecture is beginning to die out, no ISA slots,
no DIN or PS/2 keyboards, no traditional I/O ports (gate A20), no serial &
parallel ports, no traditional video hardware, no floppies. Even the CPU´s
Real Mode might be going sooner or later. On such machines, DOS may
continue to run only in emulated environments such as under DOSEMU,
not as a native OS any more.
(The kernel itself is relatively save because it has some sort of
"hardware abstraction layer" (no virtualization, of course), but the
memory managers and some of the disk tools would require massive
changes, as they are often directly accessing hardware. Although a new
"32-bit" DOS could be made compatible with existing software, the current
situation does not justify such a effort IMHO. The alternative is running
DOS under Linux, and Linux to be made much easier and flexible to
configure and use.) However, there are (and will be) still myards of
older systems using the traditional PC architecture, fast enough to
fulfill their tasks, not to speak of embedded x86 boards, which will
remain available for the long term.
- (for max. 5 years due to the mentioned hardware changes ahead) "OS
independent" boot floppy (e.g. things like Ontrack´s Easy Recovery,
Storagesoft´s ImageCast, StarTools´ disk tools etc.) and recovery BIOS
use (e.g. IBM´s DFT or new "dual-BIOS" features).
- (at least for 10 to 15 years) Compatibility box under Linux
This could evolve into some kind of general "Legacy Platform" for all kind
of ancient software (including emulators for long forgotten systems of the
"home computer" era) and important data files from the past three decades.
(With multi-user support and some handy COMMAND.COM extensions
for improved Unix support at the DOS prompt. Also interesting: Advanced
NLS sub-system to provide limited Unicode support to DOS and bring in all
the code pages found in other OSes, including the ISO, Windows, and Mac
code pages.)
Some of these paths may be commercially viable, others are definitely not,
but would still be useful for convenience, flexibility, compatibility, or
just to hold up the decreasing acceptance of DOS in the mainstream
(every user counts). However, without publically available sources, all this
is just wishful thinking (at least for DR-DOS).
Let´s hope, this year will bring some changes, and things around
DR-DOS will start moving again... (No need to be idle until than,
help supporting FreeDOS and Linux meanwhile!!! This will indirectly
also help DR-DOS.)
In closing, I´d like to wish you all good default settings... ;->
Matthias
BTW. Without being able to promise anything because of the
applying legal restrictions as well as obvious limited resources
on my end of the wire, I would personally appreciate to receive
all your detailed bug and problem reports regarding *any* issue
of DR-DOS (for example, those you might once have sent to
Caldera or Lineo), (realistic) feature and usability suggestions,
etc...
-----------------------------------------------------------------
Matthias Paul, Ubierstrasse 28, D-50321 Bruehl, Germany
<Matthias DOT Paul AT post DOT rwth-aachen DOT de> <mpaul AT drdos DOT org>
http://www.rhrz.uni-bonn.de/~uzs180/mpdokeng.html
-----------------------------------------------------------------
- Raw text -