Mail Archives: opendos/1997/03/20/08:43:09
On Thu, 13 Mar 1997, Evan Dickinson wrote:
> > > While I agree with the idea of a standard, I abhor hardcoded directory
> > > names. I've already got my own directory structure and I hate programs
> > > that won't respect that.
> >
> > Well, I've got my own directory structure too, and mine is
> > probably different from yours. That is the reason for pushing
> > for such a standard. If everyone has things in the same spot, it
> > makes life easier. Everyone will want their current heirarchy to
> > be "the standard" and none will become it. As much as some
> > people may not like the idea of such a standard, it will probably
> > come to be anyway. Hard coded directories are not necessarily
> > part of this standard however. I think that hard coded dirs
> > should not be used.
>
> First, let me see if I understand your proposition. You'd like the
> default directory for, say, utilities to be c:\util (or something). A
> user could then install to a different directory, but at a loss of
> compatibility. Correct me if I'm wrong here.
Well, not really compatibility per-se, but more like this.
A user installs a program into the standard directory, then the
install is logged in some way by the as yet to be decided OpenDOS
"smart installer" then when you want to uninstall, you just run
the OD uninstaller, and you're done with it. If you decide to
not use the standard, then you're stuck uninstalling the stuff
yourself as is now the case. Either way, you still can do
whatever you want, either abiding by the OPTIONAL standard or
not. New "smart installer" aware programs would make the install
process easier, also, they would install in the conventional way
if no smart installer was present.
> > > Different users would enter their own directories instead of mine.
> >
> > But possibly lose compatibility. Also, they'd just be continuing
> > to support the DOS chaotic directory structure. My vote is
> > definately for the standard. Once again, it won't be hard coded
> > dirs though. If you install elsewhere, then you'll have to do a
> > manual uninstall, or figure it out yourself.
>
> They wouldn't lose compatability. A program aware of this system would
> look up the app directory, and then look there for whatever files it
> needs. It wouldn't be furthering chaos because all shared libraries would
> be in lib, all utilities in a subdirectory of util. As long as you
> install into a directory that you've specified, everything will remain
> compatible.
Yes it WOULD maintain chaos. Although it would be possible to
hack the installer to use user defined directories, the OTHER
purposes of a standard would be null and void so why bother at
all? Some of the other "purposes" being that a particular
program or config file would be in the EXACT SAME SPOT on EVERY
COMPUTER that used the new standard. If a user moves it, then
that is fine, and they are responsible for breaking the standard.
Everything would still work mind you, but just not to the
standard. It would be equivalent to me saying "Oh, that /usr
directory in Linux has a dumb name, I'm going to call it
/non_essential_programs_and_files instead."
If we are going to try and make such a standard at all, we've got
to either make something USEFUL or nothing at all. When I talk
about these standards, I'm not talking about an "install
standard" I'm talking about a "filesystem standard". This
standard NAMES the standard directories where things will be
installed, etc. It is just a RECOMMENDATION however and nothing
more. What others are saying isn't such a standard at all but
more of a smart install program with no real standard at all.
Needless to say, either type of thing would be better than what
we've got, however a filesystem standard would be the superior
approach. If such a standard were created, then over time most
OpenDOS computers directory structures would look very much the
same and anyone could jump from machine to machine in DOS very
easy way without trying to do a filefind or a DIR /S, etc...
Again, since the standard would ONLY BE A RECOMMENDATION, other
users would NOT have to abide by it anyways. Those who did
however would benefit greatly by the features of the standard.
> > > Then an installer wanting to install off of an app directory would
> read > > this file, and ask if it should install into c:\dosapps,
> d:\dosapps or > > another directory. A simple "type paths.dir" would show
> you were > > everything is. > > > > This way, we keep our directories and
> enforce a standard. > >
>
> > Well, I think that some sort of comprimise
> > between the two will > be reached. It just needs to be discussed at
> > great length.
> Agreed. (Sorry about the formatting. I accidentally hit Justify in
> Pine.)
Heheheh. I do that from time to time. I wonder if ^J can be
disabled? Hmmm.
Well, I think that it's already been done now. Now we
only have to decide what all we'd like to see in a filesystem
standard, then let people vote on filenames/dirnames, etc...
The comprimise is that if someone doesn't like a decision that
was made by the standard making process, they don't have to
follow it. Nothing will break (hopefully), but they'll have to
do more manual configuring, they will be looked upon as
outsiders, they will lose all of their friends, peer pressure
will drive them to drink and lose their job, etc... :o)
Mike A. Harris | http://blackwidow.saultc.on.ca/~mharris
Computer Consultant | Coming soon: dynamic-IP-freedom...
My dynamic address: http://blackwidow.saultc.on.ca/~mharris/ip-address.html
mailto:mharris AT blackwidow DOT saultc DOT on DOT ca
Visit my homepage if you want your Dynamic IP address on your webpage.
- Raw text -