AndyC said:
No. You do this on the server to unpack the installation structure from the
compressed CD image. You only need to do it once.
Okay, you unpack and and put them in the file server.
You want ordinary users to be able to install applications on the server?
I'm talking in the workstation. As you stated earlier, Administrator
priviledge
(on workstation) is needed to enable users in the workstation able to run
the network apps, correct?
As you can see, you can push the file or ask the workstation to run a script
from AD, understood, but it overkill. All you need to do is a login script
to set up a search patch for that apps executable and dlls.
So, we do not have to spend on hardware(AD Server) and software(anoter
Windows 2003 Server OS) to just push that apps -- I think this would
simplify many things.
Per-user settings can be configured by a user as usual. *Unless* an
Administrator specifically restricts them to do otherwise.
Okay, understood.
Users files/settings are kept in their profile folder as usual. Have you
ever actually used Windows?
Yes, I have.
It's hardly trouble, it takes approximately two minutes, regardless of
whether you have 1, 10 or 10,000,000 workstations. The advantage of Active
Directory is that you do not need to ever touch the actual workstation
machines at all. They don't even need to be in the same country. Heck, they
don't even need to exist yet, they'll pick up the settings when they are
added to the Domain.
If you are talking WUS(Windows Update Services), it is not yet available
(Information Week this week).
I haven't avoided any issues. You're either misreading what I said or really
haven't got a clue how networked applications work on any OS.
Let me rephrase it again:
1. There is an Apps called: AppsA
2. Install apps A in the server --- e.g: \\servername\appsdir\AppsA\
3. UserXX in WorkstationY want to access AppsA
4. UserXX in WorkstationY type: \\<servername>\appsdir\AppsA\A.exe
5. Voila, UserXX is running AppsA in his/her WorkstationYY.
In short, install appsA in the <servername>, and then 1000 users can access
those apps
in a blink by just typing \\<servername>\appsdir\AppsA\A.exe (he/she can
create
a shortcut anytime)
When there are updates, all we need to do is update those AppsA in <
servername>\appsdir\AppsA directory, and VOILA, all users/workstation
see the update.
What! That's one step. A single command. And only necessary if you aren't
using Active Directory. How much less do you expect to do? Even a *nix based
OS needs one command to mount a networked application drive for crying out
loud.
In *nix, install apps in /usr/local/AppsA, automount /usr/local/AppsA in
user workstation, then execute the /usr/local/AppsA/A.exe ...
NOW user run the apps.
When patch comes in, update the file(s) in /usr/local/AppsA.
As soon as the patch finishes in the server, all the user can see the update
right away.
Nothing has to be run on the workstation! All the user has to do is to type
in the command line:
/usr/local/AppsA/A.exe (okay, they do not use .exe for executable
extension).
That's what I just explained (several times) how to do. Why not try reading
the posts instead of assuming you know better.
Not exactly! And by the way, I do read all your comments.
On the contrary. It's probably the one thing that Windows simply does better
than any other operating system.
Sure! I think you know that is not true! So, I would not go there.
If you really feel the need to slag it off,
there are plenty of areas where Unix/Linux/Mac actually have a genuine
advantage. If you'd ever run a large scale network, you'd know this
already.
Yes, I have.
I'm an administrator in a mixed Solaris/Linux/Mac environment, I do know
what I'm talking about.
Same here.
