[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
fwd: WHY windows project mailing list
Might want to get on the list to keep track of his
progress. Looks like it could be promising. Or maybe
just the leaders should get on the list, and keep SEUL
------- Forwarded Message
From: Perry Harrington <firstname.lastname@example.org>
Subject: WHY windows project mailing list
Date: Sun, 25 Jan 1998 22:16:44 -0800 (PST)
Due to the amount of response I've received (and the duplication of ideas I've
been writing), I have created a mailing list. If you are not interested in
being on this list, please email me, and I'll remove you.
If you know anyone else that would be interested in this, please let them
know about this project. Tell them to email email@example.com to subscribe.
Ok, the mailing list is firstname.lastname@example.org. You can email to it, it's
just a Sendmail :include: exploder list, so it works like the Linux kernel
list is that you need to CC responses to it.
A little background about me:
My name is Perry Harrington, and I started out using a Morse Telecommunications
release of Slackware 2.1. I installed linux in March of 1995 and have been
happily using it since. I work as a System Software Engineer for a web
hosting company (WebCom) in Santa Cruz California. About August of 1997,
I got interested in building a unified user interface for Linux. It started
out because I wanted to learn X by reading the "X Toolkit Cookbook", and
realized how evil the X design is. I concluded that X had 3 fundamentally
1. It was complex to develop an application because of the primitive nature
of it's programming API. X is difficult to program from the ground up. Sure,
2. It duplicates the event handling and object management of the user interface
in every program. This is inefficient because it wastes memory, kernel
bandwidth (unneccessary socket communication), and doesn't centralize the
3. It does not maintain a consistent interface between each program. This is
a major detractor to commercial acceptance of Linux...they don't have the
guarantee that their program will work the same from one machine to the next,
or that you have all the resources neccessary to run it.
More importantly, I had evaluated Caldera, Red Hat, and Slackware, and I still
find Slackware much easier to install in spite of it's shortcomings. I
decided that WHY could be used in the initial install of Linux, doing everything
right, and making it small. Ideally WHY should run from 1 or 2 1.44 floppies
for a complete Linux install (Herc/EGA/VGA support only).
So, now you all know a little bit about what sparked this project. Now, why
did I want to pursue WHY now? Well, the Netscape Communicator 5.0 source is
a big part (porting Netscape to WHY could give a big push to acceptance), the
more important part is that I read the famous "Cathedral vs Bazaar" paper.
Since I read it, I have wanted to try the concept out, hopefully I can use it
where I work (I've spent the past 2.5 years trying to get them to use Linux
in a mission critical role, and it's finally happening :) if is pans out.
So, I recognized that the Linux community is really a better collection of
programming knowledge and experience than any other place around. Hence this
project, now. I hope to get a number of people that specialize in various
aspects of GUI and object management (font people, sound people, compiler
people, and object people, etc) to assist. I don't personally have the
ability to mash fonts (anything more than bitmap) or postscript, etc.
To conclude, I hope that all of you will be the start of the most successful,
(windowing and GUI) project, next to the Linux project itself.
I will be getting together some of the core concepts and architectural ideas
pretty soon, right now they're just chicken scratches on the floor of my
Perry Harrington Linux rules all OSes. APSoft ()
email: email@example.com Think Blue. /\
------- End of Forwarded Message