[plt-scheme] Why is Planet better than PLT archives?

On Fri, 2005-08-05 at 11:47 -0500, Jacob Matthews wrote:
   . . .
> PLaneT is a distribution system; it uses PLT archives. So really the 
> question is "Why is distributing my package with PLaneT better than just 
> putting it on my web page?" The answer to which is, (a) PLaneT is 
> centralized (http://planet.plt-scheme.org) so your package gets exposed 
> more widely and (b) Mz/DrScheme automatically download appropriate 
> packages in response to the (require (planet ...)) form if they're not 
> already available on the local machine. That means that in writing 
> programs that you intend to distribute to other people, you can just 
> refer to the PLaneT packages you want and have  confidence that whoever 
> uses your program will automatically get the additional libraries they 
> need if they don't already have them.
> 
> Basically, before PLaneT, if you were writing a program for others to 
> use, you had to weigh the advantage of using a third-party library 
> against the pain of having to make others install it manually. PLaneT 
> makes you not have to do that anymore.

OK, I have some questions as a PLaneT *user*.  There's a nice package
that I want to use in my app.  My goal is to get the package installed
into my file system in a sane place, like the collects dir.

1)  Do I just write a mzscheme source file with the appropriate
    (require ...) form (and possibly a dummy expr to be evaluated) to
    get the package loaded locally?

2)  If the package consists of more than one module will the one require
    do the trick, or do I have to figure out some dummy app that
    exercises all the facilities to get everything transferred?

3)  I have PLT installed in /usr/local.  Do I have to be privileged to
    do the download/install?

In short, what are the nuts, bolts and gotchas of obtaining and using
PLaneT packages?

Thanks,

 -- Bill Wood