My vision of Debtop (complete version)
(Sorry about the previous version, I accidentally clicked send)
My vision of Debtop (a name I once heard this project called) is a complete and
user friendly operating system. Now before I elaborate on this I must warn you
that to achieve this goal much work would need to be done. Now let's get on
with it. I see Debtop as a complete and user friendly operating system from end
to end. The ideal Debtop would include a simple graphical installer with simple
partition resizing support. The system would also support a quick graphical
boot with plenty of feedback, and hardware auto detection.
The lower level stuff:
As stated above the system would have a fast graphical boot with feedback and
automatic hardware detection and configuration. Also after boot the system
would graphically communicate events such as network cables being unplugged, new
wireless hotspots, new hardware (usb, firewire) and such.
The higher level stuff:
The user would be able to change his screen resolution and graphics *driver* on
the fly with out logging out or restarting his display stuff. He (or she)
should also be able to configure multiple monitors with out restarting anything.
The user's interface would be uniform and easy to use with useful hints in
useful places. The user would be able to add and configure new hardware with
ease including devices ranging from scanners, cameras, and external drives. The
users block devices would automatically mount and unmount them selves. Selected
users will be able to install and remove selected programs from the system with
ease. The user would be able to apply security updates in a simple manner, or
have them automatically applied for him and be alerted about it. The user
should be able to use his keyboard hotkeys.
Developers for the system will have clear, simple to use APIs that support many
language bindings. Packaging software for distribution on the system should be
simple and easy.
Things that exist that could be used to this end:
apt and deb packages
GNOME & KDE
Notice X is not on the list, because fixing X would take more effort than
rewriting it. The y-windows project already has working code that can do nice
fancy things that X is lacking including a uniform API for widgets (multimedia
is planned). I am a firm believer that the y-windows project could become a
competitor (or compliment) to GNOME and KDE if they play their cards right.
(NOTE: X works fine for thin client use, but desktop use is a little different)
The design of Y can be found here. I have ordered the documents in order from
the easiest and quickest read on.
http://www.kfk4ever.com/~griffon26/Thesis/MarkThomas.html (html design doc)
(pdf design doc)
Now I know I seem wildly and stupidly ambitious (because I am) but I am a firm
believer that this can be done (look at the linux kernel). Perhpaps a few
talented, ambitious, and stupidly idealistic artists, programmers, and
developers get together with a few seasoned, fed up with desktop Linux, reality
oriented, old hat developers, artists, and developers this thing could happen.
Not saying that current technologies could not be used, but Microsoft is
rewriting windows for longhorn, why can't we do the same for desktop linux?
Matthew A. Nicholson