Quoting Derek Martin <[EMAIL PROTECTED]>:
> On Mon, 31 Jan 2000, Chester Martel wrote:
> > The biggest improvement among vendors would be an 'easier' anybody can
> > install package and/or better installation documentation. Reading this
> Honestly, I think RedHat (and maybe others) has that already, at least for
> new installs. I was able to install a standard workstation install of
> RH6.1 in under 5 minutes. For the newbie, they provide a ton of
> documentation, both in printed format and on the CD.
Yep, it's a whole world better then the old days of downloading Slackware
install floppies on a 9600 baud modem.. ;-P But in oder to get to the point
of mass acceptence, a user shouldn't have to look at the docs. I know, I know,
it's one of the worst things to say, but my mom certainly doesn't want to read
about /dev/hda1, etc.. ;-P
> > list would indicate that upgrades are not problem free. If possible,
> Admittedly, that still true, although the mainstream distributions are
> getting closer there, as well. There are a lot of files whose formats
> change between versions, and a lot of the utilites change the way they are
> configured to non-standard methods (case in point being RH's AnotherLevel
> method of configuring your X window manager). It's difficult to write
> good scripts to modify old files to use new formats, in all possible
> cases.
Closer and closer every day. The proplem I see here is, it's not really a
*Linux* problem, as much as it is a *developers* mindset problem. As helish as
it is, under Win32 *everything* uses the registry, and
adding/updating/maintaining the values contained in it is much easier. A
developer reading the registry only needs to know the key names, that it. From
there, they go through the OS to retrieve/set these values. Primarily, using a
key/value system, there is no file format to deal with, and hence, no parsing
involved.. The registry itself would be a *VERY* nice thing, if the OS did
more maintence and watchdogging of it.
> At any rate, those things are being worked on. I didn't really formulate
> my thought very well, but what I was actually thinking was what things
> need to be worked on that currently aren't? One area I think that needs a
> lot of work is documentation. There is already a lot of good
> documentation out there, but it's becoming outdated, and for newer
> packages like KDE and Gnome, documentation is a bit sparse.
More generic graphical interfaces to system management. 'Point and Click'
does have its virtues. Linuxconf and several other packages are prime examples
of this need being addressed. What I'd like to see is for Distributers, such
as RedHat, building linuxconf modules for as many packages as can be
configured, providing a unified interface to nearly everything. Linuxconf
could become the 'Control Panel' of linux.
> One of the things I'm getting at is I'd like to contribute to an open
> source project, but I'd like to be able to contribute to an area that
> doesn't have a lot of existing work going on. So I'd like to get into
> a project that is just getting started, but I'm also thinking about
> creating my own project.
That's a tough one. The one thing I've learned is, don't join something
becouse you think it's cool, or neat, but something that truely personally
interests you. Join as a tester, etc, for the cool and neat projects, but I've
seen to many times where someone gets involved in a project, loses interest,
and simply stops. While this isn't a 'bad thing' really, I think that having
developers devote their interests to something that really get them going is
the best place for their energy to be put.. Just MHO, though..
> Besides that though, I just wanted to get an idea of what people think
> Linux is missing, particularly in the context of continuing to be a viable
> platform for both servers and desktops, both in and out of the enterprise.
Covered most of that above. An easy to use, end user interface to control
their system.
> Is it really just commercial software availability and people's mind set?
> Most of the arguments I've heard people make are, IMO, not really valid
> anymore, with the advent of the latest distributions, which offer so many
> features to make using linux easy. Just some thoughts.
That's hard to say, becouse, as you say, we're prejeduce.. I guess my best
stab at it would be compatibility with other systems. I'm not so sure this is,
again, a Linux problem, as much as a software developers problem. It's great
if we can import and export to all these different formats, but untill you can
literally hand a disk over to someone, and have them 'open up' what's on that
disk datawise, people will find over and over, that they simply don't to deal
with how to save the data, and just go ahead and use Win32 and MS Office. Now,
MS Office under Linux would solve that, but I'm not so sure there's a bats
chance in HELL that's going to happen anytime soon..
---
Thomas Charron
<< Wanted: One decent sig >>
<< Preferably litle used >>
<< and stored in garage. ?>>
**********************************************************
To unsubscribe from this list, send mail to
[EMAIL PROTECTED] with the following text in the
*body* (*not* the subject line) of the letter:
unsubscribe gnhlug
**********************************************************
